Datasets:

allenai

/

SciRIFF

like 32

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: A previous work along this line is Sproat et al. (1996), which is based on weighted finite-state transducers (FSTs). We propose a unified approach that solves both problems simultaneously. We believe that the identification of OOV words should not be treated as a problem separate from word segmentation. Citation Sentence: A previous work along this line is Sproat et al. ( 1996 ) , which is based on weighted finite-state transducers ( FSTs ) . Context after the citation: Our approach is similarly motivated but is based on a different mechanism: linear mixture models. As we shall see, the models provide a more flexible framework for incorporating various kinds of lexical and statistical information. Many types of OOV words that are not covered in Sproat’s system can be dealt with in our system. The linear models we used are originally derived from linear discriminant functions widely used for pattern classification (Duda, Hart, and Stork 2001) and have been recently introduced into NLP tasks by Collins and Duffy (2001).

CompareOrContrast

acl_arc_intent_classification:train:1088

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: Most approaches rely on VerbNet (Kipper et al., 2000) and FrameNet (Baker et al., 1998) to provide associations between verbs and semantic roles, that are then mapped onto the current instance, as shown by the systems competing in semantic role labelling competitions (Carreras and Marquez, 2004; Carreras and Marquez, 2005) and also (Gildea and Jurafsky, 2002; Pradhan et al., 2005; Shi and Mihalcea, 2005). In current work on semantic relation analysis, the focus is on semantic roles – relations between verbs and their arguments. The system’s interface facilitates the expert’s task of creating and manipulating structures which represent domain concepts, and assigning them relations from a relation dictionary. Citation Sentence: Most approaches rely on VerbNet ( Kipper et al. , 2000 ) and FrameNet ( Baker et al. , 1998 ) to provide associations between verbs and semantic roles , that are then mapped onto the current instance , as shown by the systems competing in semantic role labelling competitions ( Carreras and Marquez , 2004 ; Carreras and Marquez , 2005 ) and also ( Gildea and Jurafsky , 2002 ; Pradhan et al. , 2005 ; Shi and Mihalcea , 2005 ) . Context after the citation: These systems share two ideas which make them different from the approach presented here: they all analyse verb-argument relations, and they all use machine learning or probabilistic approaches (Pradhan et al., 2005) to assign a label to a new instance. Labelling every instance relies on the same previously encoded knowledge (see (Carreras and Marquez, 2004; Carreras and Marquez, 2005) for an overview of the systems in the semantic role labelling competitions from 2004 and 2005). Pradhan et al. (2005) combine the outputs of multiple parsers to extract reliable syntactic information, which is translated into features for a machine learning experiment in assigning semantic roles. Our system analyses incrementally pairs of units coming from three syntactic levels – clause (CL), intra-clause (or verb-argument, IC), noun-phrase (NP).

Background

acl_arc_intent_classification:train:1089

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: The standard way to handle this problem is to handcraft a finite set of features which provides a sufficient summary of the unbounded history (Ratnaparkhi, 1999; Collins, 1999; Charniak, 2000). Most probability estimation methods require that there be a finite set of features on which the probability is conditioned. The main difficulty with this estimation is that the history d1,..., di_1 is of unbounded length. Citation Sentence: The standard way to handle this problem is to handcraft a finite set of features which provides a sufficient summary of the unbounded history ( Ratnaparkhi , 1999 ; Collins , 1999 ; Charniak , 2000 ) . Context after the citation: The probabilities are then assumed to be independent of all the infoimation about the history which is not captured by the chosen features. The difficulty with this approach is that the choice of features can have a large impact on the performance of the system, but it is not feasible to search the space of possible feature sets by hand. One alternative to choosing a finite set of features is to use kernel methods, which can handle unbounded 2We extended the left-corner parsing model in a few minor ways using grammar transforms. We replace Chomsky adjunction structures (i.e. structures of the form [X [X ...] [Y ...]]) with a special "modifier" link in the tree (becoming [X ... [mod Y • requiring nodes which are popped from the stack to choose between attaching with a normal link or a modifier link.

CompareOrContrast

acl_arc_intent_classification:train:109

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: Berger et al (1996) describe an efficient algorithm for accomplishing this in which approximations to Pst (TIS) are computed in parallel for all (new) features ft by holding all weights in the existing model fixed and optimizing only over a8t. A powerful strategy for using gains is to build a model iteratively by adding at each step the feature which gives the highest gain with respect to those already added. Since MEMD models are trained by finding the set of feature weights which maximizes the likelihood of the training corpus, it is natural to rate features according to how much they contribute to this likelihood. Citation Sentence: Berger et al ( 1996 ) describe an efficient algorithm for accomplishing this in which approximations to Pst ( TIS ) are computed in parallel for all ( new ) features ft by holding all weights in the existing model fixed and optimizing only over a8t . Context after the citation: However, this method requires many expensive passes over the corpus to optimize the weights for the set of features under consideration at each step, and it adds only one feature per step, so it is not practical for constructing models containing thousands of features or more. In a recent paper (Printz, 1998), Printz argues that it is usually sufficient to perform the iteration described in the previous paragraph only once, in other words that features can be ranked simply according to their gain with respect to some initial model. He also gives an algorithm for computing gains using a numerical approximation which requires only a single pass over the training corpus. I adopted Printz' method for computing MEMD gains, using the reference trigram as the initial model.

Background

acl_arc_intent_classification:train:1090

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: As suggested in Rapp & Zock (2010) this can be done by looking up the ranks of each of the four given words (i.e. the words occurring in a particular word equation) within the association vector of a translation candidate, and by multiplying these ranks. However, if we assume, for example, that we already have word equations for four languages, and all we want is to compute the translations into a fifth language, then we can simply extend our approach to what we call the product-of-ranks algorithm. So far, we always computed translations to single source words. Citation Sentence: As suggested in Rapp & Zock ( 2010 ) this can be done by looking up the ranks of each of the four given words ( i.e. the words occurring in a particular word equation ) within the association vector of a translation candidate , and by multiplying these ranks . Context after the citation: So for each candidate we obtain a product of ranks. We then assume that the candidate with the smallest product will be the best translation.3 Let us illustrate this by an example: If the given words are the variants of the word nervous in English, French, German, and Spanish, i.e. nervous, nerveux, nervös, and nervioso, and if we want to find out their translation into Italian, we would look at the association vectors of each word in our Italian target vocabulary. The association strengths in these vectors need to be inversely sorted, and in each of them we will look up the positions of our four given words. Then for each vector we compute the product of the four ranks, and finally sort the Italian vocabulary according to these products.

Motivation

acl_arc_intent_classification:train:1091

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: 32 In certain cases an extension of the constraint language with named disjunctions or contexted constraints (Maxwell and Kaplan 1989; Eisele and Dorre 1990; Griffith 1996) can be used to circumvent constraint propagation. The information common to all solutions to the interaction call is lifted up into the lexical entry and becomes available upon lexical lookup. Applying constraint propagation to the extended lexical entry of Figure 17 yields the result shown in Figure 23. Citation Sentence: 32 In certain cases an extension of the constraint language with named disjunctions or contexted constraints ( Maxwell and Kaplan 1989 ; Eisele and Dorre 1990 ; Griffith 1996 ) can be used to circumvent constraint propagation . Context after the citation: Encoding the disjunctive possibilities for lexical rule application in this way, instead of with definite clause attachments, makes all relevant lexical information available at lexical lookup. For analyses proposing infinite lexica, though, a definite clause encoding of disjunctive possibilities is still necessary and constraint propagation is indispensable for efficient processing. An entry suitable for on-the-fly application (lexical rules 1 and 2 only).

Background

acl_arc_intent_classification:train:1092

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: This article represents an extension of our previous work on unsupervised event coreference resolution (Bejan et al. 2009; Bejan and Harabagiu 2010). Citation Sentence: This article represents an extension of our previous work on unsupervised event coreference resolution ( Bejan et al. 2009 ; Bejan and Harabagiu 2010 ) . Context after the citation: In this work, we present more details on the problem of solving both withinand cross-document event coreference as well as describe a generic framework for solving this type of problem in an unsupervised way. As data sets, we consider three different resources, including our own corpus (which is the only corpus available that encodes event coreference annotations across and within documents). In the next section, we provide additional information on how we performed the annotation of this corpus. Another major contribution of this article is an extended description of the unsupervised models for solving event coreference.

Extends

acl_arc_intent_classification:train:1093

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: These include devices such as interleaving the components (McDonald 1983; Appelt 1983), backtracking on failure (Appelt 1985; Nogier 1989), allowing the linguistic component to interrogate the planner (Mann 1983; Sondheimer and Nebel 1986), and Hovy's notion of restrictive (i.e., bottom-up) planning (Hovy 1988a, 1988c). There have in fact been attempts to develop modified modular designs that allow generators to handle interactions between the components. Certainly an approach to generation that does handle these interactions would be an improvement, as long as it didn't require abandoning modularity. Citation Sentence: These include devices such as interleaving the components ( McDonald 1983 ; Appelt 1983 ) , backtracking on failure ( Appelt 1985 ; Nogier 1989 ) , allowing the linguistic component to interrogate the planner ( Mann 1983 ; Sondheimer and Nebel 1986 ) , and Hovy 's notion of restrictive ( i.e. , bottom-up ) planning ( Hovy 1988a , 1988c ) . Context after the citation: All of these approaches, though, require that potential interactions be determined either by the tactical component or by the system designer in advance. The text planning component still has no way to detect and respond to unanticipated interactions on its own initiative.5 4 Danlos still has a separate low-level "syntactic" component, but essentially all of the generator's decisions are made by the strategic component. 5 In fact, adding additional components may make the problem even worse, as decisions may then be spread across three or more separate components.

Background

acl_arc_intent_classification:train:1094

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: See, among others, (Ramakrishnan et al. 1992). Magic is a compilation technique originally developed for goal-directed bottom-up processing of logic programs. State of the art top-down processing techniques are used to deal with the remaining constraints. Citation Sentence: See , among others , ( Ramakrishnan et al. 1992 ) . Context after the citation: As shown in (Minnen, 1996) •The presented research was carried out at the University of Tubingen, Germany, as part of the Sonderforschungsbereich 340. I A more detailed discussion of various aspects of the proposed parser can be found in (Minnen, 1998). magic is an interesting technique with respect to natural language processing as it incorporates filtering into the logic underlying the grammar and enables elegant control independent filtering improvements. In this paper we investigate the selective application of magic to typed feature grammars a type of constraint-logic grammar based on Typed Feature Logic (T r; G6tz, 1995).

Background

acl_arc_intent_classification:train:1095

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: In Table 2, lem refers to the LTAG parser (Sarkar et al., 2000), ANSI C implementation of the two-phase parsing algorithm that performs the head corner parsing (van Noord, 1994) without features (phase 1), and then executes feature unification (phase 2). Table 2 shows the average parsing time with the LTAG and HPSG parsers. This result empirically attested the strong equivalence of our algorithm. Citation Sentence: In Table 2 , lem refers to the LTAG parser ( Sarkar et al. , 2000 ) , ANSI C implementation of the two-phase parsing algorithm that performs the head corner parsing ( van Noord , 1994 ) without features ( phase 1 ) , and then executes feature unification ( phase 2 ) . Context after the citation: TNT refers to the HPSG parser (Torisawa et al., 2000), C++ implementation of the two-phase parsing algorithm that performs filtering with a compiled CFG (phase 1) and then executes feature unification (phase 2). Table 2 clearly shows that the HPSG parser is significantly faster than the LTAG parser. This result implies that parsing techniques for HPSG are also beneficial for LTAG 4We eliminated 32 elementary trees because the LTAG parser cannot produce correct derivation trees with them.

CompareOrContrast

acl_arc_intent_classification:train:1096

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: Burkett and Klein (2008) and Burkett et al. (2010) focused on joint parsing and alignment. Our U-trees are learned based on STSG, which is more appropriate for tree-based translation models than SCFG. This study differs from their work because we concentrate on constructing tree structures for tree-based translation models. Citation Sentence: Burkett and Klein ( 2008 ) and Burkett et al. ( 2010 ) focused on joint parsing and alignment . Context after the citation: They utilized the bilingual Tree-bank to train a joint model for both parsing and word alignment. Cohn and Blunsom (2009) adopted a Bayesian method to infer an STSG by exploring the space of alignments based on parse trees. Liu et al. (2012) re-trained the linguistic parsers bilingually based on word alignment. Burkett and Klein (2012) utilized a transformation-based method to learn a sequence of monolingual tree transformations for translation.

CompareOrContrast

acl_arc_intent_classification:train:1097

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: In modern syntactic theories (e.g., lexical-functional grammar [LFG] [Kaplan and Bresnan 1982; Bresnan 2001; Dalrymple 2001], head-driven phrase structure grammar [HPSG] [Pollard and Sag 1994], tree-adjoining grammar [TAG] [Joshi 1988], and combinatory categorial grammar [CCG] [Ades and Steedman 1982]), the lexicon is the central repository for much morphological, syntactic, and semantic information. Citation Sentence: In modern syntactic theories ( e.g. , lexical-functional grammar [ LFG ] [ Kaplan and Bresnan 1982 ; Bresnan 2001 ; Dalrymple 2001 ] , head-driven phrase structure grammar [ HPSG ] [ Pollard and Sag 1994 ] , tree-adjoining grammar [ TAG ] [ Joshi 1988 ] , and combinatory categorial grammar [ CCG ] [ Ades and Steedman 1982 ] ) , the lexicon is the central repository for much morphological , syntactic , and semantic information . Context after the citation: * National Centre for Language Technology, School of Computing, Dublin City University, Glasnevin, Dublin 9, Ireland. E-mail: {rodonovan,mburke,acahill,josef,[email protected]. † Centre for Advanced Studies, IBM, Dublin, Ireland. Submission received: 19 March 2004; revised submission received: 18 December 2004; accepted for publication: 2 March 2005.

Background

acl_arc_intent_classification:train:1098

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: For example, when books shouldn’t be copied by hand any longer, authors took the advantage and start writing original books and evaluation – i.e. literary criticism – unlike in the previous times (Eisenstein, 1983). When Gutenberg invented the printing press and Aldo Manuzio invented the book as we know it, new forms of writings arose. Terms as ‘chapter’, ‘page’ or ‘footnote’ simply become meaningless in the new texts, or they highly change their meaning. Citation Sentence: For example , when books should n't be copied by hand any longer , authors took the advantage and start writing original books and evaluation -- i.e. literary criticism -- unlike in the previous times ( Eisenstein , 1983 ) . Context after the citation: Nowadays the use of computers for writing has drammatically changed, expecially after their interconnection via the internet, since at least the foundation of the web (Berners-Lee, 1999). For example, a ‘web page’ is more similar to an infinite canvas than a written page (McCloud, 2001). Moreover, what seems to be lost is the relations, like the texture underpinning the text itself. From a positive point of view these new forms of writing may realize the postmodernist and decostructionist dreams of an ‘opera aperta’ (open work), as Eco would define it (1962).

Background

acl_arc_intent_classification:train:1099

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: Similar to (Li et al., 2013a), our summarization system is , which consists of three key components: an initial sentence pre-selection module to select some important sentence candidates; the above compression model to generate n-best compressions for each sentence; and then an ILP summarization method to select the best summary sentences from the multiple compressed sentences. Citation Sentence: Similar to ( Li et al. , 2013a ) , our summarization system is , which consists of three key components : an initial sentence pre-selection module to select some important sentence candidates ; the above compression model to generate n-best compressions for each sentence ; and then an ILP summarization method to select the best summary sentences from the multiple compressed sentences . Context after the citation: The sentence pre-selection model is a simple supervised support vector regression (SVR) model that predicts a salience score for each sentence and selects the top ranked sentences for further processing (compression and summarization). The target value for each sentence during training is the ROUGE-2 score between the sentence and the human written abstracts. We use three common features: (1) sentence position in the document; (2) sentence length; and (3) interpolated n-gram document frequency as introduced in (Ng et al., 2012). The final sentence selection process follows the

CompareOrContrast

acl_arc_intent_classification:train:11

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Since the arguments can provide useful semantic information, the SRL is crucial to many natural language processing tasks, such as Question and Answering (Narayanan and Harabagiu 2004), Information Extraction (Surdeanu et al. 2003), and Machine Translation(Boas 2002). Typical tags include Agent, Patient, Source, etc. and some adjuncts such as Temporal, Manner, Extent, etc. The semantic roles are marked and each of them is assigned a tag which indicates the type of the semantic relation with the related predicate. Citation Sentence: Since the arguments can provide useful semantic information , the SRL is crucial to many natural language processing tasks , such as Question and Answering ( Narayanan and Harabagiu 2004 ) , Information Extraction ( Surdeanu et al. 2003 ) , and Machine Translation ( Boas 2002 ) . Context after the citation: With the efforts of many researchers (Carreras and Màrquez 2004, 2005, Moschitti 2004, Pradhan et al 2005, Zhang et al 2007), different machine learning methods and linguistics resources are applied in this task, which has made SRL task progress fast. Compared to the research on English, the research on Chinese SRL is still in its infancy stage. Previous work on Chinese SRL mainly focused on how to transplant the machine learning methods which has been successful with English, such as Sun and Jurafsky (2004), Xue and Palmer (2005) and Xue (2008). Sun and Jurafsky (2004) did the preliminary work on Chinese SRL without any large semantically annotated corpus of Chinese.

Background

acl_arc_intent_classification:train:110

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Furthermore, the availability of rich ontological resources, in the form of the Unified Medical Language System (UMLS) (Lindberg et al., 1993), and the availability of software that leverages this knowledge— MetaMap (Aronson, 2001) for concept identification and SemRep (Rindflesch and Fiszman, 2003) for relation extraction—provide a foundation for studying the role of semantics in various tasks. (NLM), which also serves as a readily available corpus of abstracts for our experiments. Information that satisfies physicians’ needs can be found in the MEDLINE database maintained by the U.S. National Library of Medicine Citation Sentence: Furthermore , the availability of rich ontological resources , in the form of the Unified Medical Language System ( UMLS ) ( Lindberg et al. , 1993 ) , and the availability of software that leverages this knowledge -- MetaMap ( Aronson , 2001 ) for concept identification and SemRep ( Rindflesch and Fiszman , 2003 ) for relation extraction -- provide a foundation for studying the role of semantics in various tasks . Context after the citation: McKnight and Srinivasan (2003) have previously examined the task of categorizing sentences in medical abstracts using supervised discriminative machine learning techniques. Building on the work of Ruch et al. (2003) in the same domain, we present a generative approach that attempts to directly model the discourse structure of MEDLINE abstracts using Hidden Markov Models (HMMs); cfXXX (Barzilay and Lee, 2004). Although our results were not obtained from the same exact collection as those used by authors of these two previous studies, comparable experiments suggest that our techniques are competitive in terms of performance, and may offer additional advantages as well. Discriminative approaches (especially SVMs) have been shown to be very effective for many supervised classification tasks; see, for example, (Joachims, 1998; Ng and Jordan, 2001).

Background

acl_arc_intent_classification:train:1100

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: Since the language generation module works in parallel with the language understanding module, utterance generation is possible even while the system is listening to user utterances and that utterance understanding is possible even while it is speaking (Nakano et al., 1999a). In typical question-answer systems, the user has the initiative when asking questions and the system has it when answering. If the system holds the initiative, the module executes the initial function of the phase. Citation Sentence: Since the language generation module works in parallel with the language understanding module , utterance generation is possible even while the system is listening to user utterances and that utterance understanding is possible even while it is speaking ( Nakano et al. , 1999a ) . Context after the citation: Thus the system can respond immediately after user pauses when the user has the initiative. When the system holds the initiative, it can immediately react to an interruption by the user because user utterances are understood in an incremental way (Dohsaka and Shimazu, 1997). The time-out function is effective in moving the dialogue forward when the dialogue gets stuck for some reason. For example, the system may be able to repeat the same question with another expression and may also be able to ask the user a more specific question.

Background

acl_arc_intent_classification:train:1101

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: This idea was expanded to include nouns and their modifiers through verb nominalizations (Chomsky, 1970; Quirk et al., 1985). Tesni`ere (1959), who proposes a grouping of verb arguments into actants and circumstances, gives a set of rules to connect specific types of actants – for example, agent or instrument – to such grammatical elements as subject, direct object, indirect object. Anecdotal support for this stance comes from the fact that the grammatical notion of case is the basis for semantic relations (Misra, 1966; Gruber, 1965; Fillmore, 1968). Citation Sentence: This idea was expanded to include nouns and their modifiers through verb nominalizations ( Chomsky , 1970 ; Quirk et al. , 1985 ) . Context after the citation: We work with sentences, clauses, phrases and words, using syntactic structures generated by a parser. Our system incrementally processes a text, and extracts pairs of text units: two clauses, a verb and each of its arguments, a noun and each of its modifiers. For each pair of units, the system builds a syntactic graph surrounding the main element (main clause, head verb, head noun). It then tries to find among the previously processed instances another main element with a matching syntactic graph.

Background

acl_arc_intent_classification:train:1102

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: For example, such schema can serve as a mean to represent translation examples, or find structural correspondences for the purpose of transfer grammar learning (Menezes & Richardson, 2001), (Aramaki et al., 2001), (Watanabe et al., 2000), (Meyers et al., 2000), (Matsumoto et al., 1993), (kaji et al., 1992), and example-base machine translation EBMT3 (Sato & Nagao, 1990), (Sato, 1991), (Richardson et al., 2001), (Al-Adhaileh & Tang, 1999). Due to these limitations, instead of investigating into the synchronization of two grammars, we propose a flexible annotation schema (i.e. Synchronous Structured String-Tree Correspondence (S-SSTC)) to realize additional power and flexibility in expressing structural correspondences at the level of language sentence pairs. Similar limitations also appear in synchronous CFGs (Harbusch & Poller,1994). Citation Sentence: For example , such schema can serve as a mean to represent translation examples , or find structural correspondences for the purpose of transfer grammar learning ( Menezes & Richardson , 2001 ) , ( Aramaki et al. , 2001 ) , ( Watanabe et al. , 2000 ) , ( Meyers et al. , 2000 ) , ( Matsumoto et al. , 1993 ) , ( kaji et al. , 1992 ) , and example-base machine translation EBMT3 ( Sato & Nagao , 1990 ) , ( Sato , 1991 ) , ( Richardson et al. , 2001 ) , ( Al-Adhaileh & Tang , 1999 ) . Context after the citation:

Background

acl_arc_intent_classification:train:1103

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: Collins and Duffy (2002) define a kernel over parse trees and apply it to re-ranking the output of a parser, but the resulting feature space is restricted by the need to compute the kernel efficiently, and the results are not as good as Collins' previous work on re-ranking using a finite set of features (Collins, 2000). feature sets, but then efficiency becomes a problem. We do not believe these transforms have a major impact on performance, but we have not currently run tests without them. Citation Sentence: Collins and Duffy ( 2002 ) define a kernel over parse trees and apply it to re-ranking the output of a parser , but the resulting feature space is restricted by the need to compute the kernel efficiently , and the results are not as good as Collins ' previous work on re-ranking using a finite set of features ( Collins , 2000 ) . Context after the citation: In this work we use a method for automatically inducing a finite set of features for representing the derivation history. The method is a form of multi-layered artificial neural network called Simple Synchrony Networks (Lane and Henderson, 2001; Henderson, 2000). The outputs of this network are probability estimates computed with a log-linear model (also known as a maximum entropy model), as is done in (Ratnaparkhi, 1999). Log-linear models have proved successful in a wide variety of applications, and are the inspiration behind one of the best current statistical parsers (Charniak, 2000).

Background

acl_arc_intent_classification:train:1104

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: method Context before the citation: 9 We only use the minimal GHKM rules (Galley et al., 2004) here to reduce the complexity of the sampler. Our final experiments verify this point and we will conduct a much detailed analysis in future. Thus, compared with the conventional TMs, we believe that our final TM would not be worse due to AEs. Citation Sentence: 9 We only use the minimal GHKM rules ( Galley et al. , 2004 ) here to reduce the complexity of the sampler . Context after the citation: Under this initial STSG, the sampler modifies the initial U-trees (initial sample) to create a series of new ones (new samples) by the Gibbs operators. Consequently, new STSGs are created based on the new U-trees simultaneously and used for the next sampling operation. Repeatedly and after a number of iterations, we can obtain the final U-trees for building translation models.

Uses

acl_arc_intent_classification:train:1105

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: FBLTAG (Vijay-Shanker, 1987; Vijay-Shanker and Joshi, 1988) is an extension of the LTAG formalism. Adjunction grafts an auxiliary tree with the root node and foot node labeled x onto an internal node of another tree with the same symbol x (Figure 4). Substitution replaces a substitution node with another initial tree (Figure 3). Citation Sentence: FBLTAG ( Vijay-Shanker , 1987 ; Vijay-Shanker and Joshi , 1988 ) is an extension of the LTAG formalism . Context after the citation: In FB-LTAG, each node in the elementary trees has a feature structure, containing grammatical constraints on the node. Figure 5 shows a result of LTAG analysis, which is described not only by derived trees (i.e., parse trees) but also by derivation trees. A derivation tree is a structural description in LTAG and represents the history of combinations of elementary trees.

Background

acl_arc_intent_classification:train:1106

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: In addition to headwords, dictionary search through the pronunciation field is available; Carter (1987) has merged information from the pronunciation and hyphenation fields, creating an enhanced phonological representation which allows access to entries by broad phonetic class and syllable structure (Huttenlocher and Zue, 1983). From the master LDOCE file, we have computed alternative indexing information, which allows access into the dictionary via different routes. While no application currently makes use of this facility, the motivation for such an approach to dictionary access comes from envisaging a parser which will operate on the basis of the on-line LDOCE; and any serious parser must be able to recognise compounds before it segments its input into separate words. Citation Sentence: In addition to headwords , dictionary search through the pronunciation field is available ; Carter ( 1987 ) has merged information from the pronunciation and hyphenation fields , creating an enhanced phonological representation which allows access to entries by broad phonetic class and syllable structure ( Huttenlocher and Zue , 1983 ) . Context after the citation: In addition, a fully flexible access system allows the retrieval of dictionary entries on the basis of constraints specifying any combination of phonetic, lexical, syntactic, and semantic information (Boguraev et al., 1987). Independently, random selection of dictionary entries is also provided to allow the testing of software on an unbiased sample.

Uses

acl_arc_intent_classification:train:1107

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Research on shallow parsing was inspired by psycholinguistics arguments (Gee and Grosjean, 1983) that suggest that in many scenarios (e.g., conversational) full parsing is not a realistic strategy for sentence processing and analysis, and was further motivated by several arguments from a natural language engineering viewpoint. Thus, over the past few years, along with advances in the use of learning and statistical methods for acquisition of full parsers (Collins, 1997; Charniak, 1997a; Charniak, 1997b; Ratnaparkhi, 1997), significant progress has been made on the use of statistical learning methods to recognize shallow parsing patterns syntactic phrases or words that participate in a syntactic relationship (Church, 1988; Ramshaw and Marcus, 1995; Argamon et al., 1998; Cardie and Pierce, 1998; Munoz et al., 1999; Punyakanok and Roth, 2001; Buchholz et al., 1999; Tjong Kim Sang and Buchholz, 2000). While earlier work in this direction concentrated on manual construction of rules, most of the recent work has been motivated by the observation that shallow syntactic information can be extracted using local information by examining the pattern itself, its nearby context and the local part-of-speech information. Citation Sentence: Research on shallow parsing was inspired by psycholinguistics arguments ( Gee and Grosjean , 1983 ) that suggest that in many scenarios ( e.g. , conversational ) full parsing is not a realistic strategy for sentence processing and analysis , and was further motivated by several arguments from a natural language engineering viewpoint . Context after the citation: First, it has been noted that in many natural language applications it is sufficient to use shallow parsing information; information such as noun phrases (NPs) and other syntactic sequences have been found useful in many large-scale language processing applications including information extraction and text summarization (Grishman, 1995; Appelt et al., 1993). Second, while training a full parser requires a collection of fully parsed sentences as training corpus, it is possible to train a shallow parser incrementally. If all that is available is a collection of sentences annotated for NPs, it can be used to produce this level of analysis. This can be augmented later if more information is available.

Background

acl_arc_intent_classification:train:1108

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Despite this, to date, there has been little work on corpus-based approaches to help-desk response automation (notable exceptions are Carmel, Shtalhaim, and Soffer 2000; Lapalme and Kosseim 2003; Bickel and Scheffer 2004; Malik, Subramaniam, and Kaushik 2007). This indicates that help-desk customers may have also become more tolerant of inaccurate or incomplete automatically generated replies, provided these replies are still relevant to their problem, and so long as the customers can follow up with a request for human-generated responses if necessary. An outcome of the recent proliferation of statistical approaches, in particular in recommender systems and search engines, is that people have become accustomed to responses that are not precisely tailored to their queries. Citation Sentence: Despite this , to date , there has been little work on corpus-based approaches to help-desk response automation ( notable exceptions are Carmel , Shtalhaim , and Soffer 2000 ; Lapalme and Kosseim 2003 ; Bickel and Scheffer 2004 ; Malik , Subramaniam , and Kaushik 2007 ) . Context after the citation: A major factor limiting this work is the dearth of corpora—help-desk e-mails tend to be proprietary and are subject to privacy issues. Further, this application lacks the kind of benchmark data sets that are used in question-answering and text summarization.2 In this article, we report on our experiments with corpus-based techniques for the automation of help-desk responses. Our study is based on a large corpus of request– response e-mail dialogues between customers and operators at Hewlett-Packard. Observations from this corpus have led us to consider several methods that implement different types of corpus-based strategies.

Background

acl_arc_intent_classification:train:1109

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: Most approaches rely on VerbNet (Kipper et al., 2000) and FrameNet (Baker et al., 1998) to provide associations between verbs and semantic roles, that are then mapped onto the current instance, as shown by the systems competing in semantic role labelling competitions (Carreras and Marquez, 2004; Carreras and Marquez, 2005) and also (Gildea and Jurafsky, 2002; Pradhan et al., 2005; Shi and Mihalcea, 2005). In current work on semantic relation analysis, the focus is on semantic roles – relations between verbs and their arguments. The system’s interface facilitates the expert’s task of creating and manipulating structures which represent domain concepts, and assigning them relations from a relation dictionary. Citation Sentence: Most approaches rely on VerbNet ( Kipper et al. , 2000 ) and FrameNet ( Baker et al. , 1998 ) to provide associations between verbs and semantic roles , that are then mapped onto the current instance , as shown by the systems competing in semantic role labelling competitions ( Carreras and Marquez , 2004 ; Carreras and Marquez , 2005 ) and also ( Gildea and Jurafsky , 2002 ; Pradhan et al. , 2005 ; Shi and Mihalcea , 2005 ) . Context after the citation: These systems share two ideas which make them different from the approach presented here: they all analyse verb-argument relations, and they all use machine learning or probabilistic approaches (Pradhan et al., 2005) to assign a label to a new instance. Labelling every instance relies on the same previously encoded knowledge (see (Carreras and Marquez, 2004; Carreras and Marquez, 2005) for an overview of the systems in the semantic role labelling competitions from 2004 and 2005). Pradhan et al. (2005) combine the outputs of multiple parsers to extract reliable syntactic information, which is translated into features for a machine learning experiment in assigning semantic roles. Our system analyses incrementally pairs of units coming from three syntactic levels – clause (CL), intra-clause (or verb-argument, IC), noun-phrase (NP).

Background

acl_arc_intent_classification:train:111

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: results are based on a corpus of movie subtitles (Tiedemann 2007), and are consequently shorter sentences, whereas the En→Es results are based on a corpus of parliamentary proceedings (Koehn 2005). Horizontal axis: average number of transferred edges per sentence. Vertical axis: percentage of transferred edges that are correct. Citation Sentence: results are based on a corpus of movie subtitles ( Tiedemann 2007 ) , and are consequently shorter sentences , whereas the En → Es results are based on a corpus of parliamentary proceedings ( Koehn 2005 ) . Context after the citation: We see in Figure 10 that for both domains, the models trained using posterior regularization perform better than the baseline model trained using EM.

Uses

acl_arc_intent_classification:train:1110

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: A statistical technique which has recently become popular for NLP is Maximum Entropy/Minimum Divergence (MEMD) modeling (Berger et al., 1996). Much recent research in SMT, eg (Garcia-Varea et al., 1998; Niessen et al., 1998; Och et al., 1999; Wang and Waibel, 1998) deals with the decoding problem, either directly or indirectly because of constraints imposed on the form of the translation model. But this comes at the cost of increased decoding complexity, because the chain rule can no longer be applied as in (1) due to the reversed direction of the translation model. Citation Sentence: A statistical technique which has recently become popular for NLP is Maximum Entropy/Minimum Divergence ( MEMD ) modeling ( Berger et al. , 1996 ) . Context after the citation: One of the main strengths of MEMD is that it allows information from different sources to be combined in a principled and effective way, so it is a natural choice for modeling p(wlh, s) In this paper, I describe a MEMD model for p(wlh, s) and compare its performance to that of an equivalent linear model. I also evaluate several different methods for MEMD feature selection, including a new algorithm due to Printz (1998). To my knowledge, this is the first application of MEMD to building a large-scale translation model, and one of the few direct comparisons between a MEMD model and an almost exactly equivalent linear mode1.2

Uses

acl_arc_intent_classification:train:1111

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: Our baseline coreference system uses the C4.5 decision tree learner (Quinlan, 1993) to acquire a classifier on the training texts for determining whether two NPs are coreferent. Citation Sentence: Our baseline coreference system uses the C4 .5 decision tree learner ( Quinlan , 1993 ) to acquire a classifier on the training texts for determining whether two NPs are coreferent . Context after the citation: Following previous work (e.g., Soon et al. (2001) and Ponzetto and Strube (2006)), we generate training instances as follows: a positive instance is created for each anaphoric NP, NPj, and its closest antecedent, NPi; and a negative instance is created for NPj paired with each of the intervening NPs, NPi+1, NPi+2, ..., NPj_1. Each instance is represented by 33 lexical, grammatical, semantic, and positional features that have been employed by highwe can see, the baseline achieves an F-measure of performing resolvers such as Ng and Cardie (2002) 57.0 and a resolution accuracy of 48.4. and Yang et al. (2003), as described below.

Uses

acl_arc_intent_classification:train:1112

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: method Context before the citation: This alignment is obtained by following the same set of rules learned from the development dataset as in (Zhang and Chai, 2009). Note that in this figure the alignment between x5 = suggests from the hypothesis and u4 = opinion from the conversation segment is a pseudo alignment, which directly maps a verb term in the hypothesis to an utterance term represented by its dialogue act. Figure 3 shows an example of alignment between the conversation terms and hypothesis terms in Example 2. Citation Sentence: This alignment is obtained by following the same set of rules learned from the development dataset as in ( Zhang and Chai , 2009 ) . Context after the citation:

Uses

acl_arc_intent_classification:train:1113

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: Developed Systems Our developed system is built on the work by Chang et al. (2013), using Constrained Latent Left-Linking Model (CL3M) as our mention-pair coreference model in the joint framework10. Baseline Systems We choose three publicly available state-of-the-art end-to-end coreference systems as our baselines: Stanford system (Lee et al., 2011), Berkeley system (Durrett and Klein, 2014) and HOTCoref system (Bj¨orkelund and Kuhn, 2014). 3.1.1 can be verified empirically on both ACE-2004 and OntoNotes-5.0 datasets. Citation Sentence: Developed Systems Our developed system is built on the work by Chang et al. ( 2013 ) , using Constrained Latent Left-Linking Model ( CL3M ) as our mention-pair coreference model in the joint framework10 . Context after the citation: When the CL3M coreference system uses gold mentions or heads, we call the system Gold; when it uses predicted mentions or heads, we call the system Predicted. The mention head candidate generation module along with mention boundary detection module can be grouped together to form a complete mention detection system, and we call it H-M-MD. We can feed the predicted mentions from H-M-MD directly into the mention-pair coref- 9No parsing information is needed at evaluation time.

Uses

acl_arc_intent_classification:train:1114

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: The best performance on the Brown corpus, a 0.2% error rate, was reported by Riley (1989), who trained a decision tree classifier on a 25-million-word corpus. The best performance on the WSJ corpus was achieved by a combination of the SATZ system (Palmer and Hearst 1997) with the Alembic system (Aberdeen et al. 1995): a 0.5% error rate. State-of-theart machine learning and rule-based SBD systems achieve an error rate of 0.8–1.5% measured on the Brown corpus and the WSJ corpus. Citation Sentence: The best performance on the Brown corpus , a 0.2 % error rate , was reported by Riley ( 1989 ) , who trained a decision tree classifier on a 25-million-word corpus . Context after the citation: In the disambiguation of capitalized words, the most widespread method is POS tagging, which achieves about a 3% error rate on the Brown corpus and a 5% error rate on the WSJ corpus, as reported in Mikheev (2000). We are not aware of any studies devoted to the identification of abbreviations with comprehensive evaluation on either the Brown corpus or the WSJ corpus. In row D of Table 4, we summarized our main results: the results obtained by the application of our SBD rule set, which uses the information provided by the DCA to capitalized word disambiguation applied together with lexical lookup (as described in Section 7.5), and the abbreviation-handling strategy, which included the guessing heuristics, the DCA, and the list of 270 abbreviations (as described in Section 6). As can be seen in the table, the performance of this system is almost indistinguishable from the best previously quoted results.

CompareOrContrast

acl_arc_intent_classification:train:1115

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: Most approaches rely on VerbNet (Kipper et al., 2000) and FrameNet (Baker et al., 1998) to provide associations between verbs and semantic roles, that are then mapped onto the current instance, as shown by the systems competing in semantic role labelling competitions (Carreras and Marquez, 2004; Carreras and Marquez, 2005) and also (Gildea and Jurafsky, 2002; Pradhan et al., 2005; Shi and Mihalcea, 2005). In current work on semantic relation analysis, the focus is on semantic roles – relations between verbs and their arguments. The system’s interface facilitates the expert’s task of creating and manipulating structures which represent domain concepts, and assigning them relations from a relation dictionary. Citation Sentence: Most approaches rely on VerbNet ( Kipper et al. , 2000 ) and FrameNet ( Baker et al. , 1998 ) to provide associations between verbs and semantic roles , that are then mapped onto the current instance , as shown by the systems competing in semantic role labelling competitions ( Carreras and Marquez , 2004 ; Carreras and Marquez , 2005 ) and also ( Gildea and Jurafsky , 2002 ; Pradhan et al. , 2005 ; Shi and Mihalcea , 2005 ) . Context after the citation: These systems share two ideas which make them different from the approach presented here: they all analyse verb-argument relations, and they all use machine learning or probabilistic approaches (Pradhan et al., 2005) to assign a label to a new instance. Labelling every instance relies on the same previously encoded knowledge (see (Carreras and Marquez, 2004; Carreras and Marquez, 2005) for an overview of the systems in the semantic role labelling competitions from 2004 and 2005). Pradhan et al. (2005) combine the outputs of multiple parsers to extract reliable syntactic information, which is translated into features for a machine learning experiment in assigning semantic roles. Our system analyses incrementally pairs of units coming from three syntactic levels – clause (CL), intra-clause (or verb-argument, IC), noun-phrase (NP).

Background

acl_arc_intent_classification:train:1116

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: These include devices such as interleaving the components (McDonald 1983; Appelt 1983), backtracking on failure (Appelt 1985; Nogier 1989), allowing the linguistic component to interrogate the planner (Mann 1983; Sondheimer and Nebel 1986), and Hovy's notion of restrictive (i.e., bottom-up) planning (Hovy 1988a, 1988c). There have in fact been attempts to develop modified modular designs that allow generators to handle interactions between the components. Certainly an approach to generation that does handle these interactions would be an improvement, as long as it didn't require abandoning modularity. Citation Sentence: These include devices such as interleaving the components ( McDonald 1983 ; Appelt 1983 ) , backtracking on failure ( Appelt 1985 ; Nogier 1989 ) , allowing the linguistic component to interrogate the planner ( Mann 1983 ; Sondheimer and Nebel 1986 ) , and Hovy 's notion of restrictive ( i.e. , bottom-up ) planning ( Hovy 1988a , 1988c ) . Context after the citation: All of these approaches, though, require that potential interactions be determined either by the tactical component or by the system designer in advance. The text planning component still has no way to detect and respond to unanticipated interactions on its own initiative.5 4 Danlos still has a separate low-level "syntactic" component, but essentially all of the generator's decisions are made by the strategic component. 5 In fact, adding additional components may make the problem even worse, as decisions may then be spread across three or more separate components.

Background

acl_arc_intent_classification:train:1117

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: The system was trained on the Penn Treebank (Marcus et al., 1993) WSJ Sections 221 and tested on Section 23 (Table 1), same as used by Magerman (1995), Collins (1997), and Ratnaparkhi (1997), and became a common testbed. Citation Sentence: The system was trained on the Penn Treebank ( Marcus et al. , 1993 ) WSJ Sections 221 and tested on Section 23 ( Table 1 ) , same as used by Magerman ( 1995 ) , Collins ( 1997 ) , and Ratnaparkhi ( 1997 ) , and became a common testbed . Context after the citation: The tasks were selected so as to demonstrate the benefit of using internal structure data for learning composite structures. We have studied the effect of noun-phrase information on learning verb phrases by setting limits on the number of embedded instances, nemb in a tile. A limit of zero emulates the flat version since learning takes place from POS tags only.

CompareOrContrast

acl_arc_intent_classification:train:1118

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Bolter (1991) was the first scholar who stressed the impact of the digital revolution to the medium of writing. Citation Sentence: Bolter ( 1991 ) was the first scholar who stressed the impact of the digital revolution to the medium of writing . Context after the citation: Terms as ‘chapter’, ‘page’ or ‘footnote’ simply become meaningless in the new texts, or they highly change their meaning. When Gutenberg invented the printing press and Aldo Manuzio invented the book as we know it, new forms of writings arose. For example, when books shouldn’t be copied by hand any longer, authors took the advantage and start writing original books and evaluation – i.e. literary criticism – unlike in the previous times (Eisenstein, 1983). Nowadays the use of computers for writing has drammatically changed, expecially after their interconnection via the internet, since at least the foundation of the web (Berners-Lee, 1999).

Background

acl_arc_intent_classification:train:1119

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: In previous work (Bachenko et al. 1986), we described an experimental text-to-speech system that determined prosodic phrasing for the Olive—Liberman synthesizer (Olive and Liberman 1985). Citation Sentence: In previous work ( Bachenko et al. 1986 ) , we described an experimental text-to-speech system that determined prosodic phrasing for the Olive -- Liberman synthesizer ( Olive and Liberman 1985 ) . Context after the citation: The system generated phrase boundaries using information derived from the syntactic structure of a sentence. While we saw significant improvements in the resulting synthesized speech, we also observed problems with the system. Often these stemmed from our assumptions that both clausal structure and predicateargument relations were important in determining prosodic phrasing. This paper reconsiders those assumptions and describes an analysis of phrasing that we believe corrects many of the problems of the earlier version.

Background

acl_arc_intent_classification:train:112

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: A similar problem is discussed in the psycholinguistics of interpretation (Sedivy et al. 1999): Interpretation is widely assumed to proceed incrementally, but vague descriptions resist strict incrementality, since an adjective in a vague description can only be fully interpreted when its comparison set is known. This means that the linguistic realization cannot start until CD is concluded, contradicting eye-tracking experiments suggesting that speakers start speaking while still scanning distractors (Pechmann 1989). This question is especially pertinent in the case of vague expressions, since gradable properties are selected last, but realized first (Section 6). Citation Sentence: A similar problem is discussed in the psycholinguistics of interpretation ( Sedivy et al. 1999 ) : Interpretation is widely assumed to proceed incrementally , but vague descriptions resist strict incrementality , since an adjective in a vague description can only be fully interpreted when its comparison set is known . Context after the citation: Sedivy and colleagues resolve this quandary by allowing a kind of revision, whereby later words allow hearers to refine their interpretation of gradable adjectives. Summarizing the situation in generation and interpretation, it is clear that the last word on incrementality has not been said.

Background

acl_arc_intent_classification:train:1120

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Much of theoretical linguistics can be formulated in a very natural manner as stating correspondences (translations) between layers of representation structures (Rambow & Satta, 1996). There is now a consensus about the fact that natural language should be described as correspondences between different levels of representation. Citation Sentence: Much of theoretical linguistics can be formulated in a very natural manner as stating correspondences ( translations ) between layers of representation structures ( Rambow & Satta , 1996 ) . Context after the citation: In this paper, a flexible annotation schema called Structured String-Tree Correspondence (SSTC) (Boitet & Zaharin, 1988) will be introduced to capture a natural language text, its corresponding abstract linguistic representation and the mapping (correspondence) between these two. The correspondence between the string and its associated representation tree structure is defined in terms of the sub-correspondence between parts of the string (substrings) and parts of the tree structure (subtrees), which can be interpreted for both analysis and generation. Such correspondence is defined in a way that is able to handle some non-standard cases (e.g. non-projective correspondence). While synchronous systems are becoming more and more popular, there is therefore a great need for formal models of corresponding different levels of representation structures.

Background

acl_arc_intent_classification:train:1121

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: In our previous work (Tomuro, 2000), we applied this method to a small subset of WordNet nouns and showed potential applicability. The lexicon is derived by a fully automatic extraction method which utilizes a clustering technique called tree-cut (Li and Abe, 1998). In this paper, we describes a lexicon organized around systematic polysemy. Citation Sentence: In our previous work ( Tomuro , 2000 ) , we applied this method to a small subset of WordNet nouns and showed potential applicability . Context after the citation: In the current work, we applied the method to all nouns and verbs in WordNet, and built a lexicon in which word senses are partitioned by systematic polysemy. We report results of comparing our lexicon with the WordNet cousins as well as the inter-annotator disagreement observed between two semantically annotated corpora: WordNet Semcor (Landes et al., 1998) and DSO (Ng and Lee, 1996). The results are quite promising: our extraction method discovered 89% of the WordNet cousins, and the sense partitions in our lexicon yielded better K values (Carletta, 1996) than arbitrary sense groupings on the agreement data. 2 The Tree-cut Technique

Extends

acl_arc_intent_classification:train:1122

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: Much of the earlier work in anaphora resolution heavily exploited domain and linguistic knowledge (Sidner 1979; Carter 1987; Rich and LuperFoy 1988; Carbonell and Brown 1988), which was difficult both to represent and to process, and which required considerable human input. Last, but not least, application-driven research in areas such as automatic abstracting and information extraction independently highlighted the importance of anaphora and coreference resolution, boosting research in this area. The drive toward corpus-based robust NLP solutions further stimulated interest in alternative and/or data-enriched approaches. Citation Sentence: Much of the earlier work in anaphora resolution heavily exploited domain and linguistic knowledge ( Sidner 1979 ; Carter 1987 ; Rich and LuperFoy 1988 ; Carbonell and Brown 1988 ) , which was difficult both to represent and to process , and which required considerable human input . Context after the citation: However, the pressing need for the development of robust and inexpensive solutions to meet the demands of practical NLP systems encouraged many researchers to move away from extensive domain and linguistic knowledge and to embark instead upon knowledge-poor anaphora resolution strategies. A number of proposals in the 1990s deliberately limited the extent to which they relied on domain and/or linguistic knowledge and reported promising results in knowledge-poor operational environments (Dagan and Itai 1990, 1991; Lappin and Leass 1994; Nasukawa 1994; Kennedy and Boguraev 1996; Williams, Harvey, and Preston 1996; Baldwin 1997; Mitkov 1996, 1998b). The drive toward knowledge-poor and robust approaches was further motivated by the emergence of cheaper and more reliable corpus-based NLP tools such as partof-speech taggers and shallow parsers, alongside the increasing availability of corpora and other NLP resources (e.g., ontologies). In fact, the availability of corpora, both raw and annotated with coreferential links, provided a strong impetus to anaphora resolu-

Background

acl_arc_intent_classification:train:1123

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: the mention sub-type, which is a sub-category of the mention type (ACE, 2004) (e.g. OrgGovernmental, FacilityPath, etc.). the mention class (generic, specific, negatively quantified, etc.) 4. the mention level (named, nominal, pronominal, or premodifier) 3. Citation Sentence: the mention sub-type , which is a sub-category of the mention type ( ACE , 2004 ) ( e.g. OrgGovernmental , FacilityPath , etc. ) . Context after the citation:

Uses

acl_arc_intent_classification:train:1124

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: For more details on the proprieties of SSTC, see Boitet & Zaharin (1988). The particle "up" is featurised into the verb "pick" and in discontinuous manner (e.g. "up" (4-5) in "pick-up" (1-2+4-5)) in the sentence "He picks the box up". The case depicted in Figure 2, describes how the SSTC structure treats some non-standard linguistic phenomena. Citation Sentence: For more details on the proprieties of SSTC , see Boitet & Zaharin ( 1988 ) . Context after the citation:

Background

acl_arc_intent_classification:train:1125

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: This has been reported for other languages, too, dependent on the generality of the chosen approach (J¨appinen and Niemist¨o, 1988; Choueka, 1990; Popovic and Willett, 1992; Ekmekc¸ioglu et al., 1995; Hedlund et al., 2001; Pirkola, 2001). mers (Lovins, 1968; Porter, 1980) demonstrably improve retrieval performance. For English, known for its limited number of inflection patterns, lexicon-free general-purpose stem1‘ ’ denotes the string concatenation operator. Citation Sentence: This has been reported for other languages , too , dependent on the generality of the chosen approach ( J ¨ appinen and Niemist ¨ o , 1988 ; Choueka , 1990 ; Popovic and Willett , 1992 ; Ekmekc ¸ ioglu et al. , 1995 ; Hedlund et al. , 2001 ; Pirkola , 2001 ) . Context after the citation: When it comes to a broader scope of morphological analysis, including derivation and composition, even for the English language only restricted, domain-specific algorithms exist. This is particularly true for the medical domain. From an IR view, a lot of specialized research has already been carried out for medical applications, with emphasis on the lexico-semantic aspects of dederivation and decomposition (Pacak et al., 1980; Norton and Pacak, 1983; Wolff, 1984; Wingert, 1985; Dujols et al., 1991; Baud et al., 1998). While one may argue that single-word compounds are quite rare in English (which is not the case in the medical domain either), this is certainly not true for German and other basically agglutinative languages known for excessive single-word nominal compounding.

Background

acl_arc_intent_classification:train:1126

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: method Context before the citation: Specifically, we used Decision Graphs (Oliver 1993) for Doc-Pred, and SVMs (Vapnik 1998) for Sent-Pred.11 Additionally, we used unigrams for clustering documents and sentences, and unigrams and bigrams for predicting document clusters and sentence clusters (Sections 3.1.2 and 3.2.2). Hence, throughout the course of this project, the different methods had minor implementational variations, which do not affect the overall insights of this research. The focus of our work is on the general applicability of the different response automation methods, rather than on comparing the performance of particular implementation techniques. Citation Sentence: Specifically , we used Decision Graphs ( Oliver 1993 ) for Doc-Pred , and SVMs ( Vapnik 1998 ) for Sent-Pred .11 Additionally , we used unigrams for clustering documents and sentences , and unigrams and bigrams for predicting document clusters and sentence clusters ( Sections 3.1.2 and 3.2.2 ) . Context after the citation: Because this variation was uniformly implemented for both approaches, it does not affect their relative performance. These methodological variations are summarized in Table 2. As indicated at the beginning of this section, the implementation of these methods requires the selection of different thresholds, which are subjective and application dependent. Table 3 summarizes the thresholds required for the different methods, the

Uses

acl_arc_intent_classification:train:1127

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: To prepare SMT outputs for post-editing, the creators of the corpus used their own WMT10 system (Potet et al., 2010), based on the Moses phrase-based decoder (Koehn et al., 2007) with dense features. The corpus is a subset of the newscommentary dataset provided at WMT4 and contains input French sentences, MT outputs, postedited outputs and English references. We used the LIG corpus3 which consists of 10,881 tuples of French-English post-edits (Potet et al., 2012). Citation Sentence: To prepare SMT outputs for post-editing , the creators of the corpus used their own WMT10 system ( Potet et al. , 2010 ) , based on the Moses phrase-based decoder ( Koehn et al. , 2007 ) with dense features . Context after the citation: We replicated a similar Moses system using the same monolingual and parallel data: a 5-gram language model was estimated with the KenLM toolkit (Heafield, 2011) on news.en data (48.65M sentences, 1.13B tokens), pre-processed with the tools from the cdec toolkit (Dyer et al., 2010). perceptron cycling theorem (Block and Levin, 1970; Gelfand et al., 2010) should suffice to show a similar bound. 3http://www-clips. imag.fr/geod/User/marion.

Uses

acl_arc_intent_classification:train:1128

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: To model o(Li,S→T), o(Ri,S→T), i.e. the reordering of the neighboring phrases of a function word, we employ the orientation model introduced by Setiawan et al. (2007). Citation Sentence: To model o ( Li , S → T ) , o ( Ri , S → T ) , i.e. the reordering of the neighboring phrases of a function word , we employ the orientation model introduced by Setiawan et al. ( 2007 ) . Context after the citation: Formally, this model takes the form of probability distribution Pori(o(Li,S→T),o(Ri,S→T)|Yi,S→T), which conditions the reordering on the lexical identity of the function word alignment (but independent of the lexical identity of its neighboring phrases). In particular, o maps the reordering into one of the following four orientation values (borrowed from Nagata et al. (2006)) with respect to the function word: Monotone Adjacent (MA), Monotone Gap (MG), Reverse Adjacent (RA) and Reverse Gap (RG). The Monotone/Reverse distinction indicates whether the projected order follows the original order, while the Adjacent/Gap distinction indicates whether the pro- 2This heuristic is commonly used in learning phrase pairs from parallel text.

Uses

acl_arc_intent_classification:train:1129

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Note that this ensures that greater importance is attributed to longer chunks, as is usual in most EBMT systems (cfXXX Sato and Nagao 1990; Veale and Way 1997; Carl 1999).7 As an example, consider the translation into French of the house collapsed. In order to calculate a ranking for each TL sentence produced, we multiply the weights of each chunk used in its construction. When translated phrases have been retrieved for each chunk of the input string, they must then be combined to produce an output string. Citation Sentence: Note that this ensures that greater importance is attributed to longer chunks , as is usual in most EBMT systems ( cfXXX Sato and Nagao 1990 ; Veale and Way 1997 ; Carl 1999 ) .7 As an example , consider the translation into French of the house collapsed . Context after the citation: Assume the conditional probabilities in (33): 7 Note that approaches that prefer the greatest context to be taken into account are not limited to EBMT. Research in the area of data-oriented parsing (cfXXX Bod, Scha, and Sima’an, 2003) also shows that unless the corpus is inherently biased, derivations constructed using the smallest number of subtrees have a higher probability than those built with a larger number of smaller subtrees. Computational Linguistics Volume 29, Number 3

Background

acl_arc_intent_classification:train:113

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: The syntactic structures of the input data are produced by a parser with good coverage and detailed syntactic information, DIPETT (Delisle and Szpakowicz, 1995). The difference between the number of types (2850) and tokens (573) in the extracted pairs (which contain only open-class words) shows that the same concepts recur, as expected in a didactic text. There are 4686 word tokens and 969 types. Citation Sentence: The syntactic structures of the input data are produced by a parser with good coverage and detailed syntactic information , DIPETT ( Delisle and Szpakowicz , 1995 ) . Context after the citation: The parser, written in Prolog, implements a classic constituency English grammar from Quirk et al. (1985). Pairs of syntactic units connected by grammatical relations are extracted from the parse trees. A dependency parser would produce a similar output, but DIPETT also provides verb subcategorization information (such as, for example, subject-verb-object or subject-verb-objectindirect object), which we use to select the (best) matching syntactic structures.

Uses

acl_arc_intent_classification:train:1130

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: In particular, (Gross, 1989) lists the converses of some 3 500 predicative nouns. For complementing this database and for converse constructions, the LADL tables (Gross, 1975) can furthermore be resorted to, which list detailed syntactico-semantic descriptions for 5 000 verbs and 25 000 verbal expressions. For shuffling paraphrases, french alternations are partially described in (Saint-Dizier, 1999) and a resource is available which describes alternation and the mapping verbs/alternations for roughly 1 700 verbs. Citation Sentence: In particular , ( Gross , 1989 ) lists the converses of some 3 500 predicative nouns . Context after the citation:

FutureWork

acl_arc_intent_classification:train:1131

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: This is similar to the “deletion” strategy employed by Zettlemoyer and Collins (2007), but we do it directly in the grammar. If no parse is found yet, then the parser attempts to strategically allow tokens to subsume a neighbor by making it a dependent (first with a restricted root set, then without). If that fails, then it searches for a parse with any root. Citation Sentence: This is similar to the `` deletion '' strategy employed by Zettlemoyer and Collins ( 2007 ) , but we do it directly in the grammar . Context after the citation: We add unary rules of the form (D)-*u for every potential supertag u in the tree. Then, at each node spanning exactly two tokens (but no higher in the tree), we allow rules t-*((D), v) and t-*(v, (D)). Recall that in §3.1, we stated that (D) is given extremely low probability, meaning that the parser will avoid its use unless it is absolutely necessary. Additionally, since u will still remain as the preterminal, it will be the category examined as the context by adjacent constituents.

CompareOrContrast

acl_arc_intent_classification:train:1132

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: Although this study falls under the general topic of discourse modeling, our work differs from previous attempts to characterize text in terms of domainindependent rhetorical elements (McKeown, 1985; Marcu and Echihabi, 2002). Nevertheless, their work bolsters our claims regarding the usefulness of generative models in extrinsic tasks, which we do not describe here. Whereas Barzilay and Lee evaluated their work in the context of document summarization, the fourpart structure of medical abstracts allows us to conduct meaningful intrinsic evaluations and focus on the sentence classification task. Citation Sentence: Although this study falls under the general topic of discourse modeling , our work differs from previous attempts to characterize text in terms of domainindependent rhetorical elements ( McKeown , 1985 ; Marcu and Echihabi , 2002 ) . Context after the citation: Our task is closer to the work of Teufel and Moens (2000), who looked at the problem of intellectual attribution in scientific texts.

CompareOrContrast

acl_arc_intent_classification:train:1133

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Many NLP applications require knowledge about semantic relatedness rather than just similarity (Budanitsky and Hirst, 2006). tween two words (Gurevych, 2005).3 Dissimilar words can be semantically related, e.g. via functional relationships (night – dark) or when they are antonyms (high – low). 2In this paper, word denotes the graphemic form of a token and concept refers to a particular sense of a word. Citation Sentence: Many NLP applications require knowledge about semantic relatedness rather than just similarity ( Budanitsky and Hirst , 2006 ) . Context after the citation: A number of competing approaches for computing semantic relatedness of words have been developed (see Section 2). A commonly accepted method for evaluating these approaches is to compare their results with a gold standard based on human judgments on word pairs. For that purpose, relatedness scores for each word pair have to be determined experimentally. Creating test datasets for such experiments has so far been a labor-intensive manual process.

Background

acl_arc_intent_classification:train:1134

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: Other work on modeling the meanings of verbs using video recognition has also begun showing great promise (Mathe et al., 2008; Regneri et al., 2013). norms. More recently, Silberer et al. (2013) show that visual attribute classifiers, which have been immensely successful in object recognition (Farhadi et al., 2009), act as excellent substitutes for feature Citation Sentence: Other work on modeling the meanings of verbs using video recognition has also begun showing great promise ( Mathe et al. , 2008 ; Regneri et al. , 2013 ) . Context after the citation: The Computer Vision community has also benefited greatly from efforts to unify the two modalities. To name a few examples, Rohrbach et al. (2010) and Socher et al. (2013) show how semantic information from text can be used to improve zero-shot classification (i.e., classifying never-before-seen objects), and Motwani and Mooney (2012) show that verb clusters can be used to improve activity recognition in videos.

Background

acl_arc_intent_classification:train:1135

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: It has been shown (Roland and Jurafsky 1998) that the subcategorization tendencies of verbs vary across linguistic domains. The Brown corpus comprises 24,242 trees compiled from a variety of text genres including popular lore, general fiction, science fiction, mystery and detective fiction, and humor. Penn-III consists of the WSJ section from Penn-II as well as a parse-annotated subset of the Brown corpus. Citation Sentence: It has been shown ( Roland and Jurafsky 1998 ) that the subcategorization tendencies of verbs vary across linguistic domains . Context after the citation: Our aim, therefore, is to increase the scope of the induced lexicon not only in terms of the verb lemmas for which there are entries, but also in terms of the frames with which they co-occur. The f-structure annotation algorithm was extended with only minor amendments to cover the parsed Brown corpus. The most important of these was the way in which we distinguish between oblique and adjunct. We noted in Section 4 that our method of assigning an oblique annotation in Penn-II was precise, albeit conservative.

Motivation

acl_arc_intent_classification:train:1136

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: In the system, we extract both the minimal GHKM rules (Galley et al., 2004), and the rules of SPMT Model 1 (Galley et al., 2006) with phrases up to length L=5 on the source side. The system is implemented based on (Galley et al., 2006) and (Marcu et al. 2006). The translation system used for testing the effectiveness of our U-trees is our in-house stringto-tree system (abbreviated as s2t). Citation Sentence: In the system , we extract both the minimal GHKM rules ( Galley et al. , 2004 ) , and the rules of SPMT Model 1 ( Galley et al. , 2006 ) with phrases up to length L = 5 on the source side . Context after the citation: We then obtain the composed rules by composing two or three adjacent minimal rules. To build the above s2t system, we first use the parse tree, which is generated by parsing the English side of the bilingual data with the Berkeley parser (Petrov et al., 2006). Then, we binarize the English parse trees using the head binarization approach (Wang et al., 2007) and use the resulting binary parse trees to build another s2t system. For the U-trees, we run the Gibbs sampler for 1000 iterations on the whole corpus.

Uses

acl_arc_intent_classification:train:1137

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Note that this ensures that greater importance is attributed to longer chunks, as is usual in most EBMT systems (cfXXX Sato and Nagao 1990; Veale and Way 1997; Carl 1999).7 As an example, consider the translation into French of the house collapsed. In order to calculate a ranking for each TL sentence produced, we multiply the weights of each chunk used in its construction. When translated phrases have been retrieved for each chunk of the input string, they must then be combined to produce an output string. Citation Sentence: Note that this ensures that greater importance is attributed to longer chunks , as is usual in most EBMT systems ( cfXXX Sato and Nagao 1990 ; Veale and Way 1997 ; Carl 1999 ) .7 As an example , consider the translation into French of the house collapsed . Context after the citation: Assume the conditional probabilities in (33): 7 Note that approaches that prefer the greatest context to be taken into account are not limited to EBMT. Research in the area of data-oriented parsing (cfXXX Bod, Scha, and Sima’an, 2003) also shows that unless the corpus is inherently biased, derivations constructed using the smallest number of subtrees have a higher probability than those built with a larger number of smaller subtrees. Computational Linguistics Volume 29, Number 3

Background

acl_arc_intent_classification:train:1138

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: One of the better-known approaches is described in Grefenstette and Tapanainen (1994), which suggested that abbreviations first be extracted from a corpus using abbreviation-guessing heuristics akin to those described in Section 6 and then reused in further processing. Not much information has been published on abbreviation identification. 12.2.3 Disambiguation of Abbreviations. Citation Sentence: One of the better-known approaches is described in Grefenstette and Tapanainen ( 1994 ) , which suggested that abbreviations first be extracted from a corpus using abbreviation-guessing heuristics akin to those described in Section 6 and then reused in further processing . Context after the citation: This is similar to our treatment of abbreviation handling, but our strategy is applied on the document rather than corpus level. The main reason for restricting abbreviation discovery to a single document is that this does not presuppose the existence of a corpus in which the current document is similar to other documents. Park and Byrd (2001) recently described a hybrid method for finding abbreviations and their definitions.

Background

acl_arc_intent_classification:train:1139

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: This approach is taken, for example, in LKB (Copestake 1992) where lexical rules are introduced on a par with phrase structure rules and the parser makes no distinction between lexical and nonlexical rules (Copestake 1993, 31). Another common approach to lexical rules is to encode them as unary phrase structure rules. Citation Sentence: This approach is taken , for example , in LKB ( Copestake 1992 ) where lexical rules are introduced on a par with phrase structure rules and the parser makes no distinction between lexical and nonlexical rules ( Copestake 1993 , 31 ) . Context after the citation: A similar method is included in PATR-II (Shieber et al. 1983) and can be used to encode lexical rules as binary relations in the CUF system (Dorre and Eisele 1991; Done and Dorna 1993b) or the TFS system (Emele and Zajac 1990; Emele 1994). The covariation approach described in this paper can be viewed as a domain-specific refinement of such a treatment of lexical rules. The encoding of lexical rules used in the covariation approach is related to the work of van Noord and Bouma (1994), who describe the hand-encoding of a single lexical rule as definite relations and show how these relations can be used to constrain a lexical entry. The covariation approach builds on this proposal and extends it in three ways: First, the approach shows how to detect and encode the interaction of a set of lexical rules.

CompareOrContrast

acl_arc_intent_classification:train:114

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: method Context before the citation: Unlike the models proposed by Brown et al. (1993b), this model is symmetric, because both word bags are generated together from a joint probability distribution. The probability distribution trans (ii, it') is a word-to-word translation model. Now, a bag-to-bag translation model can be fully specified by the distributions of 1 and trans. Citation Sentence: Unlike the models proposed by Brown et al. ( 1993b ) , this model is symmetric , because both word bags are generated together from a joint probability distribution . Context after the citation: Brown and his colleagues' models, reviewed in Section 4.3, generate one half of the bitext given the other half, so they are represented by conditional probability distributions. A sequenceto-sequence translation model can be obtained from a word-to-word translation model by combining Equation 11 with order information as in Equation 8. 3. The One-to-One Assumption

CompareOrContrast

acl_arc_intent_classification:train:1140

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: method Context before the citation: We apply two different priming experiments namely, the cross modal priming and masked priming experiment discussed in (Forster and Davis, 1984; Rastle et al., 2000;Marslen-Wilson et al., 1994; Marslen-Wilson et al., 2008) for Bangla morphologically complex words. Citation Sentence: We apply two different priming experiments namely , the cross modal priming and masked priming experiment discussed in ( Forster and Davis , 1984 ; Rastle et al. , 2000 ; Marslen-Wilson et al. , 1994 ; Marslen-Wilson et al. , 2008 ) for Bangla morphologically complex words . Context after the citation: Here, the prime is morphologically derived form of the target presented auditorily (for cross modal priming) or visually (for masked priming). The subjects were asked to make a lexical decision whether the given target is a valid word in that language. The same target word is again probed but with a different audio or visual probe called the control word. The control shows no relationship with the target.

Uses

acl_arc_intent_classification:train:1141

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: (Watanabe et al., 2007; Chiang et al., 2008; Hopkins and May, 2011) proposed other optimization objectives by introducing a margin-based and ranking-based indirect loss functions. (Och and Ney, 2002; Blunsom et al., 2008) used maximum likelihood estimation to learn weights for MT. (Och, 2003; Moore and Quirk, 2008; Zhao and Chen, 2009; Galley and Quirk, 2011) employed an evaluation metric as a loss function and directly optimized it. Several works have proposed discriminative techniques to train log-linear model for SMT. Citation Sentence: ( Watanabe et al. , 2007 ; Chiang et al. , 2008 ; Hopkins and May , 2011 ) proposed other optimization objectives by introducing a margin-based and ranking-based indirect loss functions . Context after the citation: All the methods mentioned above train a single weight for the whole development set, whereas our local training method learns a weight for each sentence. Further, our translation framework integrates the training and testing into one unit, instead of treating them separately. One of the advantages is that it can adapt the weights for each of the test sentences. Our method resorts to some translation examples, which is similar as example-based translation or translation memory (Watanabe and Sumita, 2003; He et al., 2010; Ma et al., 2011).

Background

acl_arc_intent_classification:train:1142

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Our work extends directions taken in systems such as Ariane (Vauquois and Boitet, 1985), FoG (Kittredge and Polguere, 1991), JOYCE (Rambow and Korelsky, 1992), and LFS (Iordanskaja et al., 1992). It has been used for transformations of conceptual and syntactic structures during generation in monolingual and multilingual natural language generation (NLG) and for transfer in machine translation (MT). In this paper we present a linguistically motivated framework for uniform lexicostructural processing. Citation Sentence: Our work extends directions taken in systems such as Ariane ( Vauquois and Boitet , 1985 ) , FoG ( Kittredge and Polguere , 1991 ) , JOYCE ( Rambow and Korelsky , 1992 ) , and LFS ( Iordanskaja et al. , 1992 ) . Context after the citation: Although it adopts the general principles found in the abovementioned systems, the approach presented in this paper is more practical, and we believe, would eventually integrate better with emerging statistics-based approaches to MT. * The work performed on the framework by this coauthor was done while at CoGenTex, Inc. The framework consists of a portable Java environment for building NLG or MT applications by defining modules using a core tree transduction engine and single declarative ASCII specification language for conceptual or syntactic dependency tree structures' and their transformations. Developers can define new modules, add or remove modules, or modify their connections. Because the processing of the transformation engine is restricted to transduction of trees, it is computationally efficient.

Extends

acl_arc_intent_classification:train:1143

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Since the arguments can provide useful semantic information, the SRL is crucial to many natural language processing tasks, such as Question and Answering (Narayanan and Harabagiu 2004), Information Extraction (Surdeanu et al. 2003), and Machine Translation(Boas 2002). Typical tags include Agent, Patient, Source, etc. and some adjuncts such as Temporal, Manner, Extent, etc. The semantic roles are marked and each of them is assigned a tag which indicates the type of the semantic relation with the related predicate. Citation Sentence: Since the arguments can provide useful semantic information , the SRL is crucial to many natural language processing tasks , such as Question and Answering ( Narayanan and Harabagiu 2004 ) , Information Extraction ( Surdeanu et al. 2003 ) , and Machine Translation ( Boas 2002 ) . Context after the citation: With the efforts of many researchers (Carreras and Màrquez 2004, 2005, Moschitti 2004, Pradhan et al 2005, Zhang et al 2007), different machine learning methods and linguistics resources are applied in this task, which has made SRL task progress fast. Compared to the research on English, the research on Chinese SRL is still in its infancy stage. Previous work on Chinese SRL mainly focused on how to transplant the machine learning methods which has been successful with English, such as Sun and Jurafsky (2004), Xue and Palmer (2005) and Xue (2008). Sun and Jurafsky (2004) did the preliminary work on Chinese SRL without any large semantically annotated corpus of Chinese.

Background

acl_arc_intent_classification:train:1144

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: Such systems extract information from some types of syntactic units (clauses in (Fillmore and Atkins, 1998; Gildea and Jurafsky, 2002; Hull and Gomez, 1996); noun phrases in (Hull and Gomez, 1996; Rosario et al., 2002)). In other methods, lexical resources are specifically tailored to meet the requirements of the domain (Rosario and Hearst, 2001) or the system (Gomez, 1998). Some methods of semantic relation analysis rely on predefined templates filled with information from processed texts (Baker et al., 1998). Citation Sentence: Such systems extract information from some types of syntactic units ( clauses in ( Fillmore and Atkins , 1998 ; Gildea and Jurafsky , 2002 ; Hull and Gomez , 1996 ) ; noun phrases in ( Hull and Gomez , 1996 ; Rosario et al. , 2002 ) ) . Context after the citation: Lists of semantic relations are designed to capture salient domain information. In the Rapid Knowledge Formation Project (RKF) a support system was developed for domain experts. It helps them build complex knowledge bases by combining components: events, entities and modifiers (Clark and Porter, 1997). The system’s interface facilitates the expert’s task of creating and manipulating structures which represent domain concepts, and assigning them relations from a relation dictionary.

Background

acl_arc_intent_classification:train:1145

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: Association Norms (AN) is a collection of association norms collected by Schulte im Walde et al. (2012). Citation Sentence: Association Norms ( AN ) is a collection of association norms collected by Schulte im Walde et al. ( 2012 ) . Context after the citation: In association norm experiments, subjects are presented with a cue word and asked to list the first few words that come to mind. With enough subjects and responses, association norms can provide a common and detailed view of the meaning components of cue words. After removing responses given only once in the entire study, the data set contains a total of 95,214 cue-response pairs for 1,012 nouns and 5,716 response types. Feature Norms (FN) is our new collection of feature norms for a group of 569 German nouns.

Uses

acl_arc_intent_classification:train:1146

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Stanford University is developing the English Resource Grammar, an HPSG grammar for English, as a part of the Linguistic Grammars Online (LinGO) project (Flickinger, 2000). There are a variety of works on efficient parsing with HPSG, which allow the use of HPSGbased processing in practical application contexts (Flickinger et al., 2000). The center of Figure 6 shows a rule application to “can run” and “we”. Citation Sentence: Stanford University is developing the English Resource Grammar , an HPSG grammar for English , as a part of the Linguistic Grammars Online ( LinGO ) project ( Flickinger , 2000 ) . Context after the citation: In practical context, German, English, and Japanese HPSG-based grammars are developed and used in the Verbmobil project (Kay et al., 1994). Our group has developed a wide-coverage HPSG grammar for Japanese (Mitsuishi et al., 1998), which is used in a high-accuracy Japanese dependency analyzer (Kanayama et al., 2000).

Background

acl_arc_intent_classification:train:1147

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: method Context before the citation: The changes made were inspired by those described in Stetina and Nagao (1997, page 75). Before doing the annotation, though, some preprocessing of the data was required to maximize the matching between our corpus and WordNet. Entropy is 4 The automatic annotation of nouns and verbs in the corpus has been done by matching them with the WordNet database files. Citation Sentence: The changes made were inspired by those described in Stetina and Nagao ( 1997 , page 75 ) . Context after the citation: To lemmatize the words we used “morpha,” a lemmatizer developed by John A. Carroll and freely available at the address: http://www.informatics.susx.ac.uk./research/nlp/carroll/morph.html. Upon simple observation, it showed a better performance than the frequently used Porter Stemmer for this task. a more informative measure of the dispersion of a distribution, which depends both on the range and on the shape of a distribution. The head dependence measure based on entropy, then, is calculated as indicated in equation (7), which calculates the entropy of the probability distribution generated by the random variable X, whose values are all the heads that co-occur with a given PP.

Motivation

acl_arc_intent_classification:train:1148

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: The grammar code system used in LDOCE is based quite closely on the descriptive grammatical framework of Quirk et al. (1972, 1985). Once the grammar codes have been restructured, it still remains to be shown that the information they encode is going to be of some utility for natural language processing. Citation Sentence: The grammar code system used in LDOCE is based quite closely on the descriptive grammatical framework of Quirk et al. ( 1972 , 1985 ) . Context after the citation: The codes are doubly articulated; capital letters represent the grammatical relations which hold between a verb and its arguments and numbers represent subcategorisation frames which a verb can appear in. Most of the subcategorisation frames are specified by syntactic category, but some are very ill-specified; for instance, 9 is defined as "needs a descriptive word or phrase". In practice many adverbial and predicative complements will satisfy this code, when attached to a verb; for example, put [X9] where the code marks a locative adverbial prepositional phrase vs. make under sense 14 (hereafter written make(14)) is coded [X9] where it marks a predicative noun phrase or prepositional phrase. The criteria for assignment of capital letters to verbs is not made explicit, but is influenced by the syntactic and semantic relations which hold between the verb and its arguments; for example, IS, L5 and T5 can all be assigned to verbs which take a NP subject and a sentential complement, but L5 will only be assigned if there is a fairly close semantic link between the two arguments and T5 will be used in preference to IS if the verb is felt to be semantically two place rather than one place, such as know versus appear.

Extends

acl_arc_intent_classification:train:1149

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: 27 Briscoe and Copestake (1996) argue that semi-productivity of lexical rules, which can be understood as a generalization of exceptions to lexical rules, can be integrated with our approach by assigning probabilities to the automaton associated with a particular lexical entry. The way these predicates interconnect is represented in Figure 19. Interaction predicates encoding lexical rule interaction for the natural classes of lexical entries in the lexicon. Citation Sentence: 27 Briscoe and Copestake ( 1996 ) argue that semi-productivity of lexical rules , which can be understood as a generalization of exceptions to lexical rules , can be integrated with our approach by assigning probabilities to the automaton associated with a particular lexical entry . Context after the citation: 28 In order to distinguish the different interaction predicates for the different classes of lexical entries, the compiler indexes the names of the interaction predicates. Since for expository reasons we will only discuss one kind of lexical entry in this paper, we will not show those indices in the examples given. 4. Partial Unfolding of Frame Predicates

Background

acl_arc_intent_classification:train:115

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: Our task was made possible by the fact that while far from being a database in the accepted sense of the word, the LDOCE typesetting tape is the only truly computerised dictionary of English (Michiels, 1983). Given that we were targeting all envisaged access routes from LDOCE to systems implemented in Lisp, and since the natural data structure for Lisp is the s-expression, we adopted the approach of converting the tape source into a set of list structures, one per entry. Finally, the complexity of the data structures stored on disc should not be constrained in any way by the method of access, as we do not have a very clear idea what form the restructured dictionary may eventually take. Citation Sentence: Our task was made possible by the fact that while far from being a database in the accepted sense of the word , the LDOCE typesetting tape is the only truly computerised dictionary of English ( Michiels , 1983 ) . Context after the citation: The logical structure of a dictionary entry is reflected on the tape as a sequence of typed records (see Figure 1), each with additional internal segmentation, where records and fields correspond to separate units in an entry, such as headword, pronunciation, grammar code, word senses, and so forth. The "lispification" of the typesetting tape was carried out in a series of batch jobs, via a program written in a general text editing facility. The need to carry out the conversion without any loss of information meant that special attention had to be paid to the large number of non-printing characters which appear on the tape. Most of these signal changes in the typographic appearance of the printed dictionary, where crucial information about the structure of an entry is represented by changes of typeface and font size.

Background

acl_arc_intent_classification:train:1150

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: A detailed description of the kinds of expectation mechanisms appearing in these systems appears in Fink (1983). While some of these systems did exhibit expectation capabilities at the sentence level, none acquired dialogues of the kind described here for the sake of dialogue level expectation and error correction. Most of these efforts concentrated on the interaction between low level information sources from a speech recognizer and a natural language processor to discover the meaning of an input sentence. Citation Sentence: A detailed description of the kinds of expectation mechanisms appearing in these systems appears in Fink ( 1983 ) . Context after the citation: The problem of handling ill-formed input has been studied by Carbonell and Hayes (1983), Granger (1983), Jensen et al. (1983), Kwasny and Sondheimer (1981), Riesbeck and Schank (1976), Thompson (1980), Weischedel and Black (1980), and Weischedel and Sondheimer (1983). A wide variety of techniques have been developed for addressing problems at the word, phrase, sentence, and in some cases, dialogue level. However, these methodologies have not used historical information at the dialogue level as described here. In most cases, the goal of these systems is to characterize the ill-formed input into classes of errors and to correct on that basis.

Background

acl_arc_intent_classification:train:1151

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: method Context before the citation: Using the GHKM algorithm (Galley et al. 2004), we can get two different STSG derivations from the two U-trees based on the fixed word alignment. Obviously, towards an s-node for sampling, the two values of `P would define two different U-trees. Otherwise, we change its state to the right state (`P=1), and transform the U-tree to Figure 3(b) accordingly. Citation Sentence: Using the GHKM algorithm ( Galley et al. 2004 ) , we can get two different STSG derivations from the two U-trees based on the fixed word alignment . Context after the citation: Each derivation carries a set of STSG rules (i.e., minimal GHKM translation rules) of its own. In the two derivations, the STSG rules defined by the two states include the one rooted at the s-node’s lowest ancestor frontier node, and the one rooted at the s-node if it is a frontier node. For instance, in Figure 3(a), as the s-node is not a frontier node, the left state (`P=0) defines only one rule: Differently, in Figure 3(b), the s-node is a frontier node and thus the right state (`P=1) defines two rules:

Uses

acl_arc_intent_classification:train:1152

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: The need for information systems to support physicians at the point of care has been well studied (Covell et al., 1985; Gorman et al., 1994; Ely et al., 2005). For a variety of reasons, medicine is an interesting domain of research. This paper presents experiments with generative content models for analyzing the discourse structure of medical abstracts, which has been confirmed to follow the four-section pattern discussed above (Salanger-Meyer, 1990). Citation Sentence: The need for information systems to support physicians at the point of care has been well studied ( Covell et al. , 1985 ; Gorman et al. , 1994 ; Ely et al. , 2005 ) . Context after the citation: Retrieval techniques can have a large impact on how physicians access and leverage clinical evidence. Information that satisfies physicians’ needs can be found in the MEDLINE database maintained by the U.S. National Library of Medicine (NLM), which also serves as a readily available corpus of abstracts for our experiments. Furthermore, the availability of rich ontological resources, in the form of the Unified Medical Language System (UMLS) (Lindberg et al., 1993), and the availability of software that leverages this knowledge— MetaMap (Aronson, 2001) for concept identification and SemRep (Rindflesch and Fiszman, 2003) for relation extraction—provide a foundation for studying the role of semantics in various tasks.

Background

acl_arc_intent_classification:train:1153

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Task properties Determining whether or not a speaker supports a proposal falls within the realm of sentiment analysis, an extremely active research area devoted to the computational treatment of subjective or opinion-oriented language (early work includes Wiebe and Rapaport (1988), Hearst (1992), Sack (1994), and Wiebe (1994); see Esuli (2006) for an active bibliography). Note that from an experimental point of view, this is a very convenient problem to work with because we can automatically determine ground truth (and thus avoid the need for manual annotation) simply by consulting publicly available voting records. In this paper, we investigate the following specific instantiation of this problem: we seek to determine from the transcripts of U.S. Congressional floor debates whether each “speech” (continuous single-speaker segment of text) represents support for or opposition to a proposed piece of legislation. Citation Sentence: Task properties Determining whether or not a speaker supports a proposal falls within the realm of sentiment analysis , an extremely active research area devoted to the computational treatment of subjective or opinion-oriented language ( early work includes Wiebe and Rapaport ( 1988 ) , Hearst ( 1992 ) , Sack ( 1994 ) , and Wiebe ( 1994 ) ; see Esuli ( 2006 ) for an active bibliography ) . Context after the citation: In particular, since we treat each individual speech within a debate as a single “document”, we are considering a version of document-level sentiment-polarity classification, namely, automatically distinguishing between positive and negative documents (Das and Chen, 2001; Pang et al., 2002; Turney, 2002; Dave et al., 2003). Most sentiment-polarity classifiers proposed in the recent literature categorize each document independently. A few others incorporate various measures of inter-document similarity between the texts to be labeled (Agarwal and Bhattacharyya, 2005; Pang and Lee, 2005; Goldberg and Zhu, 2006). Many interesting opinion-oriented documents, however, can be linked through certain relationships that occur in the context of evaluative discussions.

Background

acl_arc_intent_classification:train:1154

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Table look-up using an explicit translation lexicon is sufficient and preferable for many multilingual NLP applications, including "crummy" MT on the World Wide Web (Church & Hovy, 1993), certain machine-assisted translation tools (e.g. (Macklovitch, 1994; Melamed, 1996b)), concordancing for bilingual lexicography (Catizone et al., 1993; Gale & Church, 1991), computerassisted language learning, corpus linguistics (Melby. However, the IBM models, which attempt to capture a broad range of translation phenomena, are computationally expensive to apply. Over the past decade, researchers at IBM have developed a series of increasingly sophisticated statistical models for machine translation (Brown et al., 1988; Brown et al., 1990; Brown et al., 1993a). Citation Sentence: Table look-up using an explicit translation lexicon is sufficient and preferable for many multilingual NLP applications , including `` crummy '' MT on the World Wide Web ( Church & Hovy , 1993 ) , certain machine-assisted translation tools ( e.g. ( Macklovitch , 1994 ; Melamed , 1996b ) ) , concordancing for bilingual lexicography ( Catizone et al. , 1993 ; Gale & Church , 1991 ) , computerassisted language learning , corpus linguistics ( Melby . Context after the citation: 1981), and cross-lingual information retrieval (Oard & Dorr, 1996). In this paper, we present a fast method for inducing accurate translation lexicons. The method assumes that words are translated one-to-one. This assumption reduces the explanatory power of our model in comparison to the IBM models, but, as shown in Section 3.1, it helps us to avoid what we call indirect associations, a major source of errors in other models.

Background

acl_arc_intent_classification:train:1155

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: method Context before the citation: Our HDP extension is also inspired from the Bayesian model proposed by Haghighi and Klein (2007). We present an extension of the hierarchical Dirichlet process (HDP) model which is able to represent each observable object (i.e., event mention) by a finite number of feature types L. Citation Sentence: Our HDP extension is also inspired from the Bayesian model proposed by Haghighi and Klein ( 2007 ) . Context after the citation: However, their model is strictly customized for entity coreference resolution, and therefore, extending it to include additional features for each observable object is a challenging task (Ng, 2008; Poon and Domingos, 2008). In the HDP model, a Dirichlet process (DP) (Ferguson, 1973) is associated with each document, and each mixture component (i.e., event) is shared across documents. To describe its extension, we consider Z the set of indicator random variables for indices of events, φz the set of parameters associated with an event z, φ a notation for all model parameters, and X a notation for all random variables that represent observable features.2 Given a document collection annotated with event mentions, the goal is to find the best assignment of event indices Z*, which maximize the posterior probability P(Z|X). In a Bayesian approach, this probability is computed by integrating out all model parameters:

Motivation

acl_arc_intent_classification:train:1156

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: The EDR has close ties to the named entity recognition (NER) and coreference resolution tasks, which have been the focus of several recent investigations (Bikel et al., 1997; Miller et al., 1998; Borthwick, 1999; Mikheev et al., 1999; Soon et al., 2001; Ng and Cardie, 2002; Florian et al., 2004), and have been at the center of evaluations such as: MUC-6, MUC-7, and the CoNLL'02 and CoNLL'03 shared tasks. In this paper we focus on the Entity Detection and Recognition task (EDR) for Arabic as described in ACE 2004 framework (ACE, 2004). These tasks have applications in summarization, information retrieval (one can get all hits for Washington/person and not the ones for Washington/state or Washington/city), data mining, question answering, language understanding, etc. Citation Sentence: The EDR has close ties to the named entity recognition ( NER ) and coreference resolution tasks , which have been the focus of several recent investigations ( Bikel et al. , 1997 ; Miller et al. , 1998 ; Borthwick , 1999 ; Mikheev et al. , 1999 ; Soon et al. , 2001 ; Ng and Cardie , 2002 ; Florian et al. , 2004 ) , and have been at the center of evaluations such as : MUC-6 , MUC-7 , and the CoNLL '02 and CoNLL '03 shared tasks . Context after the citation: Usually, in computational linguistics literature, a named entity is an instance of a location, a person, or an organization, and the NER task consists of identifying each of these occurrences. Instead, we will adopt the nomenclature of the Automatic Content Extraction program (NIST, 2004): we will call the instances of textual references to objects/abstractions mentions, which can be either named (e.g. John Mayor), nominal (the president) or pronominal (she, it). An entity is the aggregate of all the mentions (of any level) which refer to one conceptual entity. For instance, in the sentence

Background

acl_arc_intent_classification:train:1157

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: method Context before the citation: As has been previously observed and exploited in the NLP literature (Pang and Lee, 2004; Agarwal and Bhattacharyya, 2005; Barzilay and Lapata, 2005), the above optimization function, unlike many others that have been proposed for graph or set partitioning, can be solved exactly in an provably efficient manner via methods for finding minimum cuts in graphs. A minimum-cost assignment thus represents an optimum way to classify the speech segments so that each one tends not to be put into the class that the individual-document classifier disprefers, but at the same time, highly associated speech segments tend not to be put in different classes. where c(s) is the “opposite” class from c(s). Citation Sentence: As has been previously observed and exploited in the NLP literature ( Pang and Lee , 2004 ; Agarwal and Bhattacharyya , 2005 ; Barzilay and Lapata , 2005 ) , the above optimization function , unlike many others that have been proposed for graph or set partitioning , can be solved exactly in an provably efficient manner via methods for finding minimum cuts in graphs . Context after the citation: In our view, the contribution of our work is the examination of new types of relationships, not the method by which such relationships are incorporated into the classification decision.

CompareOrContrast

acl_arc_intent_classification:train:1158

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: To combine the phrasal matching scores obtained at each n-gram level, and optimize their relative weights, we trained a Support Vector Machine classifier, SVMlight (Joachims, 1999), using each score as a feature. The phrasal matching score for each n-gram level is calculated as follows: Once matching for each n-gram level has been concluded, the number of matches (Mn) and the number of phrases in the hypothesis (Nn) are used to estimate the portion of phrases in H that are matched at each level (n). Citation Sentence: To combine the phrasal matching scores obtained at each n-gram level , and optimize their relative weights , we trained a Support Vector Machine classifier , SVMlight ( Joachims , 1999 ) , using each score as a feature . Context after the citation:

Uses

acl_arc_intent_classification:train:1159

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: conclusion Context before the citation: Results from other systems show that measures of semantic coherence between a student and a system were positively associated with higher learning gain (Ward and Litman, 2006). The analysis of the data we have collected indicates that student satisfaction may be affected if the system rephrases student answers using different words (for example, using better terminology) but doesn’t explicitly explain the reason why different terminology is needed (Dzikovska et al., 2010). In dialogue management and generation, the key issue we are planning to investigate is that of linguistic alignment. Citation Sentence: Results from other systems show that measures of semantic coherence between a student and a system were positively associated with higher learning gain ( Ward and Litman , 2006 ) . Context after the citation: Using a deep generator to automatically generate system feedback gives us a level of control over the output and will allow us to devise experiments to study those issues in more detail. From the point of view of tutoring research, we are planning to use the system to answer questions about the effectiveness of different approaches to tutoring, and the differences between human-human and human-computer tutoring. Previous comparisons of human-human and humancomputer dialogue were limited to systems that asked short-answer questions (Litman et al., 2006; Ros´e and Torrey, 2005). Having a system that allows more unrestricted language input will provide a more balanced comparison.

FutureWork

acl_arc_intent_classification:train:116

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: Against the background of a growing interest in multilingual NLP, multilingual anaphora /coreference resolution has gained considerable momentum in recent years (Aone and McKee 1993; Azzam, Humphreys, and Gaizauskas 1998; Harabagiu and Maiorano 2000; Mitkov and Barbu 2000; Mitkov 1999; Mitkov and Stys 1997; Mitkov, Belguith, and Stys 1998). The last decade of the 20th century saw a number of anaphora resolution projects for languages other than English such as French, German, Japanese, Spanish, Portuguese, and Turkish. The inclusion of the coreference task in the Sixth and Seventh Message Understanding Conferences (MUC-6 and MUC-7) gave a considerable impetus to the development of coreference resolution algorithms and systems, such as those described in Baldwin et al. (1995), Gaizauskas and Humphreys (1996), and Kameyama (1997). Citation Sentence: Against the background of a growing interest in multilingual NLP , multilingual anaphora / coreference resolution has gained considerable momentum in recent years ( Aone and McKee 1993 ; Azzam , Humphreys , and Gaizauskas 1998 ; Harabagiu and Maiorano 2000 ; Mitkov and Barbu 2000 ; Mitkov 1999 ; Mitkov and Stys 1997 ; Mitkov , Belguith , and Stys 1998 ) . Context after the citation: Other milestones of recent research include the deployment of probabilistic and machine learning techniques (Aone and Bennett 1995; Kehler 1997; Ge, Hale, and Charniak 1998; Cardie and Wagstaff 1999; the continuing interest in centering, used either in original or in revised form (Abracos and Lopes 1994; Strube and Hahn 1996; Hahn and Strube 1997; Tetreault 1999); and proposals related to the evaluation methodology in anaphora resolution (Mitkov 1998a, 2001b). For a more detailed survey of the state of the art in anaphora resolution, see Mitkov (forthcoming). The papers published in this issue reflect the major trends in anaphora resolution in recent years. Some of them describe approaches that do not exploit full syntactic knowledge (as in the case of Palomar et al.'s and Stuckardt's work) or that employ machine learning techniques (Soon, Ng, and Lim); others present centering-based pronoun resolution (Tetreault) or discuss theoretical centering issues (Kibble).

Background

acl_arc_intent_classification:train:1160

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: For an overview of systems designed to answer open-domain factoid questions, the TREC QA track overview papers are a good place to start (Voorhees and Tice 1999). In this section, however, we will attempt to draw connections to other clinical information systems (although not necessarily for question answering) and related domain-specific question-answering systems. As a result, there exist relatively few points of comparison to our own work, as the research space is sparsely populated. Citation Sentence: For an overview of systems designed to answer open-domain factoid questions , the TREC QA track overview papers are a good place to start ( Voorhees and Tice 1999 ) . Context after the citation: In addition, there has been much work on the application of linguistic and semantic knowledge to information retrieval; see Lin and Demner-Fushman (2006a) for a brief overview. The idea that clinical information systems should be sensitive to the practice of evidence-based medicine is not new. Based on analyses of 4,000 MEDLINE citations, Mendonc¸a and Cimino (2001) have studied MeSH terms associated with the four basic clinical tasks of therapy, diagnosis, prognosis, and etiology. The goal was to automatically classify citations for task-specific retrieval, similar in spirit to the Hedges Project (Haynes et al. 1994; Wilczynski, McKibbon, and Haynes 2001).

Background

acl_arc_intent_classification:train:1161

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: conclusion Context before the citation: Another interesting extension is to broaden the definition of a "word" to include multi-word lexical units (Smadja, 1992). Promising features for classification include part of speech, frequency of co-occurrence, relative word position, and translational entropy (Melamed, 1997). Even better accuracy can be achieved with a more fine-grained link class structure. Citation Sentence: Another interesting extension is to broaden the definition of a `` word '' to include multi-word lexical units ( Smadja , 1992 ) . Context after the citation: If such units can be identified a priori, their translations can be estimated without modifying the word-to-word model. In this manner, the model can account for a wider range of translation phenomena.

FutureWork

acl_arc_intent_classification:train:1162

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: For example, 10 million words of the American National Corpus (Ide et al., 2002) will have manually corrected POS tags, a tenfold increase over the Penn Treebank (Marcus et al., 1993), currently used for training POS taggers. Some of this new data will be manually annotated. NLP is experiencing an explosion in the quantity of electronic text available. Citation Sentence: For example , 10 million words of the American National Corpus ( Ide et al. , 2002 ) will have manually corrected POS tags , a tenfold increase over the Penn Treebank ( Marcus et al. , 1993 ) , currently used for training POS taggers . Context after the citation: This will require more efficient learning algorithms and implementations. However, the greatest increase is in the amount of raw text available to be processed, e.g. the English Gigaword Corpus (Linguistic Data Consortium, 2003). Recent work (Banko and Brill, 2001; Curran and Moens, 2002) has suggested that some tasks will benefit from using significantly more data. Also, many potential applications of NLP will involve processing very large text databases.

Background

acl_arc_intent_classification:train:1163

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: From an IR view, a lot of specialized research has already been carried out for medical applications, with emphasis on the lexico-semantic aspects of dederivation and decomposition (Pacak et al., 1980; Norton and Pacak, 1983; Wolff, 1984; Wingert, 1985; Dujols et al., 1991; Baud et al., 1998). This is particularly true for the medical domain. When it comes to a broader scope of morphological analysis, including derivation and composition, even for the English language only restricted, domain-specific algorithms exist. Citation Sentence: From an IR view , a lot of specialized research has already been carried out for medical applications , with emphasis on the lexico-semantic aspects of dederivation and decomposition ( Pacak et al. , 1980 ; Norton and Pacak , 1983 ; Wolff , 1984 ; Wingert , 1985 ; Dujols et al. , 1991 ; Baud et al. , 1998 ) . Context after the citation: While one may argue that single-word compounds are quite rare in English (which is not the case in the medical domain either), this is certainly not true for German and other basically agglutinative languages known for excessive single-word nominal compounding. This problem becomes even more pressing for technical sublanguages, such as medical German (e.g., ‘Blut druck mess gerdt’ translates to ‘device for measuring blood pressure’). The problem one faces from an IR point of view is that besides fairly standardized nominal compounds, which already form a regular part of the sublanguage proper, a myriad of ad hoc compounds are formed on the fly which cannot be anticipated when formulating a retrieval query though they appear in relevant documents. Hence, enumerating morphological variants in a semi-automatically generated lexicon, such as proposed for French (Zweigenbaum et al., 2001), turns out to be infeasible, at least for German and related languages.

Background

acl_arc_intent_classification:train:1164

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: This approach has occasionally been taken, as in Kantrowitz and Bates (1992) and Danlos (1987) and, at least implicitly, in Paris and Scott (1994) and Delin et al. (1994); however, under this approach, all of the flexibility and simplicity of modular design is lost. One possible response would be to abandon the separation; the generator could be a single component that handles all of the work. On the other hand, it precludes making decisions involving interactions between text planning and linguistic issues. Citation Sentence: This approach has occasionally been taken , as in Kantrowitz and Bates ( 1992 ) and Danlos ( 1987 ) and , at least implicitly , in Paris and Scott ( 1994 ) and Delin et al. ( 1994 ) ; however , under this approach , all of the flexibility and simplicity of modular design is lost . Context after the citation: The opposite approach is to simply ignore the limitations of a modular design and proceed as if there need be no interactions between the components. Whatever problems result will be handled as best they can, on a case-by-case basis. This approach is the one taken (implicitly or explicitly) in the majority of generators. In fact, Reiter has even argued in favor of this approach, claiming that the interactions are sufficiently minor to be ignored (or at least handled on an ad hoc basis) (Reiter 1994).

Background

acl_arc_intent_classification:train:1165

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: Typical examples are Bulgarian (Simov et al., 2005; Simov and Osenova, 2003), Chinese (Chen et al., 2003), Danish (Kromann, 2003), and Swedish (Nilsson et al., 2005). If we start by considering languages with a labeled attachment score of 85% or higher, they are characterized by high precision and recall for root nodes, typically 95/90, and by a graceful degradation of attachment score as arcs grow longer, typically 95–90–85, for arcs of length 1, 2 and 3–6. before we turn to Swedish and Turkish, focusing on recall and precision of root nodes, as a reflection of global syntactic structure, and on attachment score as a function of arc length. Citation Sentence: Typical examples are Bulgarian ( Simov et al. , 2005 ; Simov and Osenova , 2003 ) , Chinese ( Chen et al. , 2003 ) , Danish ( Kromann , 2003 ) , and Swedish ( Nilsson et al. , 2005 ) . Context after the citation: Japanese (Kawata and Bartels, 2000), despite a very high accuracy, is different in that attachment score drops from 98% to 85%, as we go from length 1 to 2, which may have something to do with the data consisting of transcribed speech with very short utterances. A second observation is that a high proportion of non-projective structures leads to fragmentation in the parser output, reflected in lower precision for roots. This is noticeable for German (Brants et al., 2002) and Portuguese (Afonso et al., 2002), which still have high overall accuracy thanks to very high attachment scores, but much more conspicuous for Czech (B¨ohmov´a et al., 2003), Dutch (van der Beek et al., 2002) and Slovene (Dˇzeroski et al., 2006), where root precision drops more drastically to about 69%, 71% and 41%, respectively, and root recall is also affected negatively. On the other hand, all three languages behave like high-accuracy languages with respect to attachment score.

Background

acl_arc_intent_classification:train:1166

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: In this paper, we use the Constrained Latent Left-Linking Model (CL3M) described in Chang et al. (2013) in our experiments. The introduction of ILP methods has influenced the coreference area too (Chang et al., 2011; Denis and Baldridge, 2007). Cardie, 2002; Bengtson and Roth, 2008; Soon et al., 2001). Citation Sentence: In this paper , we use the Constrained Latent Left-Linking Model ( CL3M ) described in Chang et al. ( 2013 ) in our experiments . Context after the citation: The task of mention detection is closely related to Named Entity Recognition (NER). Punyakanok and Roth (2001) thoroughly study phrase identification in sentences and propose three different general approaches. They aim to learn several different local classifiers and combine them to optimally satisfy some global constraints. Cardie and Pierce (1998) propose to select certain rules based on a given corpus, to identify base noun phrases.

Uses

acl_arc_intent_classification:train:1167

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: We use a standard split of 268 training documents, 68 development documents, and 106 testing documents (Culotta et al., 2007; Bengtson and Roth, 2008). Datasets The ACE-2004 dataset contains 443 documents. Citation Sentence: We use a standard split of 268 training documents , 68 development documents , and 106 testing documents ( Culotta et al. , 2007 ; Bengtson and Roth , 2008 ) . Context after the citation: The OntoNotes-5.0 dataset, which is released for the CoNLL-2012 Shared Task (Pradhan et al., 2012), contains 3,145 annotated documents. These documents come from a wide range of sources which include newswire, bible, transcripts, magazines, and web blogs. We report results on the test documents for both datasets. The ACE-2004 dataset is annotated with both mention and mention heads, while the OntoNotes5.0 dataset only has mention annotations.

Uses

acl_arc_intent_classification:train:1168

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: Aside from the extraction of theory-neutral subcategorization lexicons, there has also been work in the automatic construction of lexical resources which comply with the principles of particular linguistic theories such as LTAG, CCG, and HPSG (Chen and Vijay-Shanker 2000; Xia 1999; Hockenmaier, Bierner, and Baldridge 2004; Nakanishi, Miyao, and Tsujii 2004). Given these facts, research on automating acquisition of dictionaries for lexically based NLP systems is a particularly important issue. Manning (1993) argues that, aside from missing domain-specific complementation trends, dictionaries produced by hand will tend to lag behind real language use because of their static nature. Citation Sentence: Aside from the extraction of theory-neutral subcategorization lexicons , there has also been work in the automatic construction of lexical resources which comply with the principles of particular linguistic theories such as LTAG , CCG , and HPSG ( Chen and Vijay-Shanker 2000 ; Xia 1999 ; Hockenmaier , Bierner , and Baldridge 2004 ; Nakanishi , Miyao , and Tsujii 2004 ) . Context after the citation: In this article we present an approach to automating the process of lexical acquisition for LFG (i.e., grammatical-function-based systems). However, our approach also generalizes to CFG category-based approaches. In LFG, subcategorization requirements are enforced through semantic forms specifying which grammatical functions are required by a particular predicate. Our approach is based on earlier work on LFG semantic form extraction (van Genabith, Sadler, and Way 1999) and recent progress in automatically annotating the Penn-II and Penn-III Treebanks with LFG f-structures (Cahill et al. 2002; Cahill, McCarthy, et al. 2004).

Background

acl_arc_intent_classification:train:1169

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: These automatic transformations are based on linguistic rules (Bohmova, 2001). During the tectogrammatical parsing of Czech, the analytical tree structure is converted into the tectogrammatical one. Citation Sentence: These automatic transformations are based on linguistic rules ( Bohmova , 2001 ) . Context after the citation: Subsequently, tectogrammatical functors are assigned by the C4.5 classifier (2abokrtsk9 et al., 2002).

Uses

acl_arc_intent_classification:train:117

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: An alternative representation based on Liberman and Prince (1977) is presented in Selkirk (1984), which contends that prosody, including prosodic phrasing, is more properly represented as a grid instead of a tree. Following G&G, we require that the prosody rules build a binary tree whose terminals are phonological words and whose node labels are indices that mark boundary salience. Citation Sentence: An alternative representation based on Liberman and Prince ( 1977 ) is presented in Selkirk ( 1984 ) , which contends that prosody , including prosodic phrasing , is more properly represented as a grid instead of a tree . Context after the citation: Although a grid may be more descriptively suitable for some aspects of prosody (for example, Sproat and Liberman (1987) use the grid representation for their implementation of stress assignment in compound nominals), we are not aware of any evidence for or against a grid representation of discourseneutral phrasing. Figure 8 shows the phonological phrase tree that is built from the syntactic structure of Figure 7. The rules for building this tree apply from left to right, following the analysis we described in the preceding section. Figures 9-11.

CompareOrContrast

acl_arc_intent_classification:train:1170

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: For the joint segmentation and POS-tagging task, we present a novel solution using the framework in this article, and show that it gives comparable accuracies to our previous work (Zhang and Clark 2008a), while being more than an order of magnitude faster. For the segmentation task, we also compare our beam-search framework with alternative decoding algorithms including an exact dynamic-programming method, showing that the beam-search method is significantly faster with comparable accuracy. We give an updated set of results, plus a number of additional experiments which probe further into the advantages and disadvantages of our framework. Citation Sentence: For the joint segmentation and POS-tagging task , we present a novel solution using the framework in this article , and show that it gives comparable accuracies to our previous work ( Zhang and Clark 2008a ) , while being more than an order of magnitude faster . Context after the citation: In Section 7 we provide further discussion of the framework based on the studies of the individual tasks. We present the main advantages of the framework, and give an analysis of the main reasons for the high speeds and accuracies achieved. We also discuss how this framework can be applied to a potential new task, and show that the comparability of candidates in the incremental process is an important factor to consider. In summary, we study a general framework for incremental structural prediction, showing how the framework can be tailored to a range of syntactic processing problems to produce results competitive with the state-of-the-art.

CompareOrContrast

acl_arc_intent_classification:train:1171

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: According to the data available from 1990 U.S. Census Bureau, only 90,000 different names are shared by 100 million people (Artiles et al., 2005). A study of the query log of the AllTheWeb and Altavista search sites gives an idea of the relevance of the people search task: 11-17% of the queries were composed of a person name with additional terms and 4% were identified as person names (Spink et al., 2004). com are examples of sites which perform web people search, although with limited disambiguation capabilities. Citation Sentence: According to the data available from 1990 U.S. Census Bureau , only 90,000 different names are shared by 100 million people ( Artiles et al. , 2005 ) . Context after the citation: As the amount of information in the WWW grows, more of these people are mentioned in different web pages. Therefore, a query for a common name in the Web will usually produce a list of results where different people are mentioned. This situation leaves to the user the task of finding the pages relevant to the particular person he is interested in. The user might refine the original query with additional terms, but this risks excluding relevant documents in the process.

Background

acl_arc_intent_classification:train:1172

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: None Context before the citation: Since then this idea has been applied to several tasks, including word sense disambiguation (Yarowsky 1995) and named-entity recognition (Cucerzan and Yarowsky 1999). Gale, Church, and Yarowsky (1992) showed that words strongly tend to exhibit only one sense in a document or discourse (“one sense per discourse”). It has been applied not only to the identification of proper names, as described in this article, but also to their classification (Mikheev, Grover, and Moens 1998). Citation Sentence: Since then this idea has been applied to several tasks , including word sense disambiguation ( Yarowsky 1995 ) and named-entity recognition ( Cucerzan and Yarowsky 1999 ) . Context after the citation: Gale, Church, and Yarowsky’s observation is also used in our DCA, especially for the identification of abbreviations. In capitalized-word disambiguation, however, we use this assumption with caution and first apply strategies that rely not just on single words but on words together with their local contexts (n-grams). This is similar to “one sense per collocation” idea of Yarowsky (1993). The description of the EAGLE workbench for linguistic engineering (Baldwin et al. 1997) mentions a case normalization module that uses a heuristic in which a capitalized word in an ambiguous position should be rewritten without capitalization if it is found lower-cased in the same document.

Background

acl_arc_intent_classification:train:1173

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: introduction Context before the citation: • History-based feature models for predicting the next parser action (Black et al., 1992). • A deterministic algorithm for building labeled projective dependency graphs (Nivre, 2006). Our methodology for performing this task is based on four essential components: Citation Sentence: • History-based feature models for predicting the next parser action ( Black et al. , 1992 ) . Context after the citation: • Support vector machines for mapping histories to parser actions (Kudo and Matsumoto, 2002). • Graph transformations for recovering nonprojective structures (Nivre and Nilsson, 2005). All experiments have been performed using MaltParser (Nivre et al., 2006), version 0.4, which is made available together with the suite of programs used for preand post-processing.1 1www.

Uses

acl_arc_intent_classification:train:1174

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: method Context before the citation: For each co-occurring pair of word types u and v, these likelihoods are initially set proportional to their co-occurrence frequency („,v) and inversely proportional to their marginal frequencies n(u) and n(v) 1, following (Dunning, 1993)2. L(u, v) represents the likelihood that u and v can be mutual translations. The two hidden parameters are the probabilities of the model generating true and false positives in the data. Citation Sentence: For each co-occurring pair of word types u and v , these likelihoods are initially set proportional to their co-occurrence frequency ( „ , v ) and inversely proportional to their marginal frequencies n ( u ) and n ( v ) 1 , following ( Dunning , 1993 ) 2 . Context after the citation: When the L(u, v) are re-estimated, the model's hidden parameters come into play. After initialization, the model induction algorithm iterates: 1. Find a set of "links" among word tokens in the bitext, using the likelihood ratios and the competitive linking algorithm.

Uses

acl_arc_intent_classification:train:1175

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: experiments Context before the citation: However, rather than output this wrong translation directly, we use a post hoc validation and (if required) correction process based on Grefenstette (1999). The system simply attaches the translation with the highest weight to the existing chunk ordinateurs personnels to produce the mistranslation in (50): (50) *la ordinateurs personnels The problem of boundary friction is clearly visible here: We have inserted a feminine singular determiner into a chunk that was generalized from a masculine plural NP. The system searches for marker words within the string and retrieves their translations.10 In this case, the marker word in the string is the and its translation can be one of le, la, l’, or les, depending on the context. Citation Sentence: However , rather than output this wrong translation directly , we use a post hoc validation and ( if required ) correction process based on Grefenstette ( 1999 ) . Context after the citation: Grefenstette shows that the Web can be used as a filter on translation quality simply by searching for competing translation candidates and selecting the one that is found most often. Rather than search for competing candidates, we select the “best” translation and have its morphological variants searched for on-line. In the example above, namely, the personal computers, we search for les ordinateurs personnels versus the wrong alternatives le/la/l’ordinateurs personnels. Interestingly, using Lycos, and setting the search language to French, the correct form les ordinateurs personnels is uniquely preferred over the other alternatives, as it is found 2,454 times, whereas the others are not found at all.

Uses

acl_arc_intent_classification:train:1176

You will be presented with a citation segment from the section of an NLP research paper, as well as the context surrounding that citation. Classify the intent behind this citation by choosing from one of the following categories: - Background: provides context or foundational information related to the topic. - Extends: builds upon the cited work. - Uses: applies the methods or findings of the cited work. - Motivation: cites the work as inspiration or rationale for the research. - CompareOrContrast: compares or contrasts the cited work with others. - FutureWork: cites the work as a direction for future research. Your answer should be a single word from the following list of options: ["Background", "Extends", "Uses", "Motivation", "CompareOrContrast", "FutureWork"]. Do not include any other text in your response. Section Title: related work Context before the citation: Most approaches rely on VerbNet (Kipper et al., 2000) and FrameNet (Baker et al., 1998) to provide associations between verbs and semantic roles, that are then mapped onto the current instance, as shown by the systems competing in semantic role labelling competitions (Carreras and Marquez, 2004; Carreras and Marquez, 2005) and also (Gildea and Jurafsky, 2002; Pradhan et al., 2005; Shi and Mihalcea, 2005). In current work on semantic relation analysis, the focus is on semantic roles – relations between verbs and their arguments. The system’s interface facilitates the expert’s task of creating and manipulating structures which represent domain concepts, and assigning them relations from a relation dictionary. Citation Sentence: Most approaches rely on VerbNet ( Kipper et al. , 2000 ) and FrameNet ( Baker et al. , 1998 ) to provide associations between verbs and semantic roles , that are then mapped onto the current instance , as shown by the systems competing in semantic role labelling competitions ( Carreras and Marquez , 2004 ; Carreras and Marquez , 2005 ) and also ( Gildea and Jurafsky , 2002 ; Pradhan et al. , 2005 ; Shi and Mihalcea , 2005 ) . Context after the citation: These systems share two ideas which make them different from the approach presented here: they all analyse verb-argument relations, and they all use machine learning or probabilistic approaches (Pradhan et al., 2005) to assign a label to a new instance. Labelling every instance relies on the same previously encoded knowledge (see (Carreras and Marquez, 2004; Carreras and Marquez, 2005) for an overview of the systems in the semantic role labelling competitions from 2004 and 2005). Pradhan et al. (2005) combine the outputs of multiple parsers to extract reliable syntactic information, which is translated into features for a machine learning experiment in assigning semantic roles. Our system analyses incrementally pairs of units coming from three syntactic levels – clause (CL), intra-clause (or verb-argument, IC), noun-phrase (NP).

Background

acl_arc_intent_classification:train:1177