| { | |
| "paper_id": "W98-0106", | |
| "header": { | |
| "generated_with": "S2ORC 1.0.0", | |
| "date_generated": "2023-01-19T06:04:48.853756Z" | |
| }, | |
| "title": "Can the TAG derivation tree represent a semantic graph? An ans wer in the light of Meaning\u2022 Text Theory", | |
| "authors": [ | |
| { | |
| "first": "Marie-Helene", | |
| "middle": [], | |
| "last": "Candito", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "Universile Paris", | |
| "location": { | |
| "addrLine": "7 2, place Jussieu", | |
| "postCode": "7003, 75251", | |
| "settlement": "Paris cedex 05 marie" | |
| } | |
| }, | |
| "email": "" | |
| }, | |
| { | |
| "first": "Sylvain", | |
| "middle": [], | |
| "last": "Kahane", | |
| "suffix": "", | |
| "affiliation": { | |
| "laboratory": "", | |
| "institution": "Universile Paris", | |
| "location": { | |
| "addrLine": "7 2, place Jussieu", | |
| "postCode": "7003, 75251", | |
| "settlement": "Paris cedex 05 marie" | |
| } | |
| }, | |
| "email": "" | |
| } | |
| ], | |
| "year": "", | |
| "venue": null, | |
| "identifiers": {}, | |
| "abstract": "", | |
| "pdf_parse": { | |
| "paper_id": "W98-0106", | |
| "_pdf_hash": "", | |
| "abstract": [], | |
| "body_text": [ | |
| { | |
| "text": "From the parsing point of view, the derivation tree in TAG [hereafter DT] is seen as the \"history\" of the derivation but also as a linguistic representation, closer to semantics, that can be the basis of a further analysis. Because in TAG the elementary trees are lexicalized and localize \u2022 the predicate-arguments relations, several works have compared the DT to a structure involving dependencies between lexical items (RJ92; RVW95). 1 We agree with these authors that there are divergences between the DT and syntactic dependencies, but we show here that the DT -in the sense of (SS94) -can be viewed as a semantic dependency graph, namely a Sems for Meaning-Text Theory (MTI] (ZM67; M88) . This requires the predicate-argument cooccurrence principle and also constraints on the adjunction of prcdicative auxiliary Lrees. We briefly introduce the representation levels in MTI before studying the dependencies shown by the DT. 2", | |
| "cite_spans": [ | |
| { | |
| "start": 436, | |
| "end": 437, | |
| "text": "1", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 674, | |
| "end": 679, | |
| "text": "(MTI]", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 680, | |
| "end": 686, | |
| "text": "(ZM67;", | |
| "ref_id": "BIBREF14" | |
| }, | |
| { | |
| "start": 687, | |
| "end": 691, | |
| "text": "M88)", | |
| "ref_id": "BIBREF8" | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Introduction", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "MTI distinguishes between linguistic representations and correspondance rules to go from a representation to another, at an adjacent level. For a wr\u00fcten sentence, there are 5 representations, each with a central structure : semantic [SemS] , deep and surface syntactic [DSyntS and SSyntS] , deep and surface morphological [DMorphS and SMorphS] . At each Ievel, additional structures may supplement the central structure. A key feature of MTI is that it distinguishes between semantic and syntactic dependencies. The Sems is a graph showing semantic dependencies between semantemes (= semantic units). The dependencies are numbered to distinguish between the different 1 (RJ92) relate the DT to the deep syntactic structure (DSyntS) of MTI, namely a syntactic dependency tree, but they note thut this correspondence DT ! DSyntS is not direct, bccause the interpretation of adjunction arcs in terms of dependencies is not constant. (RVW95) take this divergence between DT and dependency tree as one of the motivations for defining D-Tree Grammars. 2 We are thankful to Anne Abei\\le, Laurence Danlos and Owen Rambow for valuable comments on earlier versions of this work.", | |
| "cite_spans": [ | |
| { | |
| "start": 233, | |
| "end": 239, | |
| "text": "[SemS]", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 269, | |
| "end": 288, | |
| "text": "[DSyntS and SSyntS]", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 322, | |
| "end": 343, | |
| "text": "[DMorphS and SMorphS]", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1046, | |
| "end": 1047, | |
| "text": "2", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Representation levels in MTT", | |
| "sec_num": "1." | |
| }, | |
| { | |
| "text": "arguments of a predicative Semanteme. An atl<litionul structure (the Sem-CommS) in<licatcs communicative features (theme-rheme. focus ... ). Figure 1 shows an example of SemS for : Tue dictionary encodes for each generalize<l lcxcmc the associated semanteme -;long with the correspondence between Sem argumenls an<l DSynt arguments. Notation : the word library is a form of thc lcxcmc LIBRARY whose semanteme is 'library'.", | |
| "cite_spans": [], | |
| "ref_spans": [ | |
| { | |
| "start": 141, | |
| "end": 149, | |
| "text": "Figure 1", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "21", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "( l)", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "21", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "We assume the following linguistic rrorcnics for elementary trees. The elementary trees correspond to exactly one semantic unit (A9 I ) 3 , and respect the predicate-argument co-coccurrence principle (P ACP), though with a semantic interprctation : scmantic predicates anchor trees with positions for the syntactic expression of all and 011/y their semantic argumcnts. 4 These positions are typed as substitution nodes and foot nodes. For instance in the tree for an attributive adjcctive, the adjective semantically governs the semanteme represented by the foot node. 5 Traditionally auxiliary trees arc used for recursive structures. If syntactic structure is considered though, another dichotomy cuts across the distinction initial/auxiliary: the syntactic hend is either the main anchor (for predicative trees) or the foot node (for modijier trees) ( K89, (SS94)). 6 All initial trees are predicative. Typical predicative auxilia\"ry trees are the trees for bridge verbs. 7", | |
| "cite_spans": [ | |
| { | |
| "start": 569, | |
| "end": 570, | |
| "text": "5", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The DT nodes as semantemes", | |
| "sec_num": "2." | |
| }, | |
| { | |
| "text": "Let us now compare DT nodes with SemS nodes. The DT refer to lexicalized elementary trees, which correspond to a semantic unit (cf supra). Therefore, a DT node can be conceived as a semanteme, plus inforrna\u00fcon for a parcicular lcxicalization of that semanteme and for a particular syntactic construction. Yet with respect to SemS nodes, two differences appear. First. in the DT, there can be several nodes in coreference (though this coreference is not handled by the TAG forrnalism), that would be represented by a single node in the Sems. And second, semantic units realized in the language as confused with the notion of semantic head. In \u00ab white car \u00bb white semantically govems car, yet car is the semantic head (a white car is a car). Following (P90) we define the semantic head as the semanteme that summarizes a semantic sub-graph. Not all sub-graphs can be summarized. In general a semantic graph for a wholc sentence does not have a single semantic head, but one for its theme and one for its rheme. 6 We follow the terminology of (SS94). Here predicative is used with its syntactic meaning. 7 Another example is the trce for glass of in a glass of wi11e. The anchor glass is the syntactic head of the whole tree (A93) . Yet the semantic interpretation of thc trees for a bridge verb and for glass-oj differ crucially: from the semantic point of view glass of behaves as a modifier and is not the semantic head of glass of wi11e. In want to stay, which expresses a will, the syntactic' head is walll.", | |
| "cite_spans": [ | |
| { | |
| "start": 750, | |
| "end": 755, | |
| "text": "(P90)", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1009, | |
| "end": 1010, | |
| "text": "6", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1222, | |
| "end": 1227, | |
| "text": "(A93)", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The DT nodes as semantemes", | |
| "sec_num": "2." | |
| }, | |
| { | |
| "text": "inflections (eg. number, tense ... ) are represented as features in TAG and, thus do not appear as nodes of the DT. So provided inflectional semantemcs are not taken into account and coreferent nodes in the DT arc considered a single node, thcre is a one-to-one relation between the SemS nodes and thc DT nodcs.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "The DT nodes as semantemes", | |
| "sec_num": "2." | |
| }, | |
| { | |
| "text": "As we said previously, several works have noted divergences between syntactic dependcncies and DT arcs. Our claim is that a constant interpretation of thc DT arcs can be found, though in terms of semwrtic and not syntactic dependencies : substituiion and adjunction arcs both reprcsent semantic dependencies, though in the opposite direction <Fig.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "DT arcs as semantic dependencies", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "3). 8 For illustration see Fig. 4 Bul obviously, the predicative adjunction arcs and the modifier adjunction arcs do not behave in the same way with respect to sy11tactic dependencies. Typically modifiers show a semantic and syntactic dependency in the opposite direction, while comp!ement auxiliary tree preserve the direction of dependency in the semantic-syntax interface. The interaction of lhe various links can cause differences between the DT and the DSyntS. Another example of mismatch is shown Fig. 4 . The DT for sentence ( l) shows the right chain of semantic dependencies for the sequence think-need-book, as the SemS shows. The only difference is the extra node for that in the DT, which does not count as a semantic unit. On the contrary in the DSyntS (Fig. 2) , a syntactic dependency appears between BOOK and THINK, without a corresponding semantic dependency.", | |
| "cite_spans": [], | |
| "ref_spans": [ | |
| { | |
| "start": 27, | |
| "end": 33, | |
| "text": "Fig. 4", | |
| "ref_id": "FIGREF2" | |
| }, | |
| { | |
| "start": 503, | |
| "end": 509, | |
| "text": "Fig. 4", | |
| "ref_id": "FIGREF2" | |
| }, | |
| { | |
| "start": 766, | |
| "end": 774, | |
| "text": "(Fig. 2)", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "DT arcs as semantic dependencies", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "'own' /0, 1 2 'librarx.V \"' 0 'book' / 1 / \\ ..... l'i 0 0 /. d' \\ 'new\"definite' 0 nee 0 \u2022 1 \"'- 'definite' ! ~. 'think' 0 1 'Mary' 1 i 'Peter'", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "DT arcs as semantic dependencies", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "So, we have seen that in the general case, a DT induces a Sems. Further, the DT contains an additional information since it detines a partial order on its nodes, so that it fonn a tree. Thus the DT defines a path to cover all nodes once. The TAG procedure, from a generation point of view, is equivalent to fixing a starting node, the DT root. From that root, semantic dependencies gone through from the govemor to the dependent (= positively) give substitution arcs, and semantic dependencies gone through in the opposite direction (= negatively) give adjunction arcs. lt can be noted that it types the elemenlary trees invo!ved as initial/auxiliary. For example, in Fig. 1 , if we want to represent 'own' as a verb with two nominal arguments extended by substitution, the structure for 'think' will necessarily be an auxiliary tree, since one of its lcaving arc has to be gone through negatively. Thus this gives another proof that bridge verbs have to be represented by auxiliary trees in relative clauses (or embedded interrogative clauses). For the same reasons, to derive (4) hhn knows the city in ll'hich Mary met Peter and read the DT as a semantic graph (see thc corresponding Sems Fig. S ),", | |
| "cite_spans": [], | |
| "ref_spans": [ | |
| { | |
| "start": 668, | |
| "end": 674, | |
| "text": "Fig. 1", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 1191, | |
| "end": 1197, | |
| "text": "Fig. S", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "DT arcs as semantic dependencies", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "if the arguments of know are to be substituted, then in has to adjoin on city and met 10 substilute in in,", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "DT arcs as semantic dependencies", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "though met is the syntactic governor of i11.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "23", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "'know' 'in' 0 0 / ) / \\ / \"'\\ / \"'\\ 'meef 0 0 0 ) / \\\"\" 0 0 'John' 'city' 'Mary'", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "23", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "'Peter' For (Sa) , in the classic TAG analysis (X9S), thc two bridge verbs adjoin recursively, und thc DT is perfect (with the interpretation of adjunction arcs dcfincd in Fig. 3 ). Yet for (Sb) Mary seems to adore fwtdog .r serves as argument for claims, but herc seems adjoins on VP, and thus claims has to adjoin on adore. 11 Thus the DT does not show the right dependencies (either semantic or syntactic, cf (R VW9S)). For (Sc), the verb surprised traditionnally receives its subject via substitution (to block extraction), thus if the bridge verb wanted is still adjoined, the DT is different from the Sems (Fig. 6) (apart from the spliuing of the 'Paul' node inw 2 coreferent nodes: we show the coreference with a curved dashed line). The problem arises because the 1ree cxstay substitutes in cxsurprise, but when the predicati ve tree \u00dfwant adjoins on cxstay, it becomes the semantic head of the whole subtree. 12 'surprise' Figure 6 : Problematic derivation (SemS and DT) for That Paul wanted to stay surprised Mary So 10 read a DT as a Sems. we need not only the PACP, but also a control ovcr the combination of the elementary trees : it must be checked that the argumental positions in a tree are actually filled by the right arguments.13 lt can be noted that for sentence (Sb) and (Sc), ru!ing out adjunctions of complement trees (as in DTG (RVW95)) solves the problem. Yet it might be problematic for sentence ( 1), for which we have seen that the TAG DT shows the right semantic dependencies. And it also ru\\es out the adjunction of an athematic complement tree (such as the one for glass-of). This is investigated in (CK98).", | |
| "cite_spans": [ | |
| { | |
| "start": 8, | |
| "end": 16, | |
| "text": "For (Sa)", | |
| "ref_id": null | |
| } | |
| ], | |
| "ref_spans": [ | |
| { | |
| "start": 172, | |
| "end": 178, | |
| "text": "Fig. 3", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 612, | |
| "end": 620, | |
| "text": "(Fig. 6)", | |
| "ref_id": null | |
| }, | |
| { | |
| "start": 932, | |
| "end": 940, | |
| "text": "Figure 6", | |
| "ref_id": null | |
| } | |
| ], | |
| "eq_spans": [], | |
| "section": "23", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "/o\" 'wrun) 2 \"\\_ 0 0 / \\ \u2022Mary' ) ~ \u2022 \" ' --f --0 'Paul' 'siay' 0 1 \\ ' / asurprise ,,.o, ' 0 aMary ,,.o /' \u00df Wan( ' -0 a Pnul", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "23", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "We have shown that in the general case the DT can be viewed as a semantic representation, in the sense of MTI, provided coreference is not taken into 11 (SS94) a!ready noted that multiple adjunctions of bridge verbs at one node should be ruled out, here we find that this ho!ds for a whole trce. 12 (K89) already noted t!iat \u00ab derivations under which thematic roles, once established, are altered by further adjunctions \u00bb should be ruled out. u Another case where positions \u00ab are not filled by the right arguments \u00bb is for instance pied-piping. The XTAG derivation for the woma11 wlwse daughter Peter talks to does not show the right semantic dependencies, since a link appears between talks-to and woman.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "Conclusion", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "Thus elementary trees can have several lexical anchors, either because some are semantically empty (empty prepositions, complementizers ... ), or because the several anchors form an idiom, whose semantic is not compositional.4 This counts for expressed semantic arguments only, so not for the agent in agentless passive constructions for instance. 5 The notion of semantic governor must not be", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| }, | |
| { | |
| "text": "In c3se of adjunctivn, the interpretution in tl!rn1!'i of", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "", | |
| "sec_num": null | |
| } | |
| ], | |
| "back_matter": [ | |
| { | |
| "text": "account. We have given a characterization of problematic derivations. This result is of crucinl importance for any further processing based on th~ TAG derivation tree. We have also provided a new characterization nf adjunction and substitution arcs depen<ling on thc direction of the semantic dependency they represent.", | |
| "cite_spans": [], | |
| "ref_spans": [], | |
| "eq_spans": [], | |
| "section": "24", | |
| "sec_num": null | |
| } | |
| ], | |
| "bib_entries": { | |
| "BIBREF0": { | |
| "ref_id": "b0", | |
| "title": "Une grammaire lexicalist!e d'arbres adjoillts pour le fram;ais", | |
| "authors": [ | |
| { | |
| "first": ")", | |
| "middle": [ | |
| "A" | |
| ], | |
| "last": "Abeille", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": null, | |
| "venue": "", | |
| "volume": "991", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "l) A. Abeille, I 991: Une grammaire lexicalist!e d'arbres adjoillts pour le fram;ais. Ph.D. Thesis. Univ. Paris 7.", | |
| "links": null | |
| }, | |
| "BIBREF1": { | |
| "ref_id": "b1", | |
| "title": "Internetions Syntaxe-Semantique dans une TAG. Colloque ILN'93", | |
| "authors": [ | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Abeille", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1993, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "A. Abeille, 1993: Internetions Syntaxe- Semantique dans une TAG. Colloque ILN'93. Nantes.", | |
| "links": null | |
| }, | |
| "BIBREF2": { | |
| "ref_id": "b2", | |
| "title": "forthcoming: Clitic Climbing and the power of TAG", | |
| "authors": [ | |
| { | |
| "first": "T", | |
| "middle": [], | |
| "last": "Bleam", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": null, | |
| "venue": "Tree-adjoi11i11g Grammars", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "T. Bleam, forthcoming: Clitic Climbing and the power of TAG, in Abeille, A\" Rambow. 0. (eds.), Tree-adjoi11i11g Grammars, CSLI. Stanford.", | |
| "links": null | |
| }, | |
| "BIBREF3": { | |
| "ref_id": "b3", | |
| "title": "Defining DTG derivations to get semantic graphs", | |
| "authors": [ | |
| { | |
| "first": "M-H", | |
| "middle": [], | |
| "last": "Candito", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "S", | |
| "middle": [], | |
| "last": "Kahane", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1998, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "M-H. Candito, S. Kahane, 1998: Defining DTG derivations to get semantic graphs. This volume.", | |
| "links": null | |
| }, | |
| "BIBREF4": { | |
| "ref_id": "b4", | |
| "title": "TAG : a formalism for text generation", | |
| "authors": [ | |
| { | |
| "first": "L", | |
| "middle": [], | |
| "last": "Danlos", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": null, | |
| "venue": "Tree-adjoi11i11g Grammars, CSLI", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "L. Danlos, forthcoming, G-TAG : a formalism for text generation, in Abeille, A., Rambow. 0. (eds.), Tree-adjoi11i11g Grammars, CSLI, Stanford.", | |
| "links": null | |
| }, | |
| "BIBREF5": { | |
| "ref_id": "b5", | |
| "title": "Sy11tactic locality a11d Tree Adjoining Grammar: Granmratical, Ac?1tisitio~1 alll~ Processing Perpectives", | |
| "authors": [ | |
| { | |
| "first": "R", | |
| "middle": [], | |
| "last": "Frank", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1992, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "R. Frank, 1992: Sy11tactic locality a11d Tree Adjoining Grammar: Granmratical, Ac?1tisitio~1 alll~ Processing Perpectives. Ph.D. Thesis. Univ. ot Pennsylvania.", | |
| "links": null | |
| }, | |
| "BIBREF6": { | |
| "ref_id": "b6", | |
| "title": "Asymmetries in Long-Distance Extraction in a Tree-Adjoining Grammar", | |
| "authors": [ | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Kroch", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1989, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "A. Kroch, 1989: Asymmetries in Long-Distance Extraction in a Tree-Adjoining Grammar. In", | |
| "links": null | |
| }, | |
| "BIBREF7": { | |
| "ref_id": "b7", | |
| "title": "Altemative Co11ceptio11s of phrase structure", | |
| "authors": [], | |
| "year": null, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Altemative Co11ceptio11s of phrase structure. M. Baltin and A. Kroch (eds), Univ. of Chicago Press, Chicago.", | |
| "links": null | |
| }, | |
| "BIBREF8": { | |
| "ref_id": "b8", | |
| "title": "Depende11cy Symax : Tlieory and Practice. Albany. State Univ", | |
| "authors": [ | |
| { | |
| "first": "I", | |
| "middle": [], | |
| "last": "", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1988, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "I. Mel'cuk, 1988: Depende11cy Symax : Tlieory and Practice. Albany. State Univ. of NY Press.", | |
| "links": null | |
| }, | |
| "BIBREF9": { | |
| "ref_id": "b9", | |
| "title": "Str11cturatio11 et mise e11 je11 procidurale d'u11 modele li11g11istiq11e declara~if dans un cadre de ge11eratio11 de texte", | |
| "authors": [ | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Polguhe", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1990, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "A. Polguhe, 1990 : Str11cturatio11 et mise e11 je11 procidurale d'u11 modele li11g11istiq11e declara~if dans un cadre de ge11eratio11 de texte. Ph.D. Thesis. Univ. of Montreal.", | |
| "links": null | |
| }, | |
| "BIBREF10": { | |
| "ref_id": "b10", | |
| "title": "A formal Look at Dependency Grammars and Phrase-Structure Grammars, with Special Consideration of Word-Order Phenomena", | |
| "authors": [ | |
| { | |
| "first": "O", | |
| "middle": [], | |
| "last": "Rambow", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Joshi", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1992, | |
| "venue": "Rece111 Trends i11 Meaning-Text Theory", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "O. Rambow, A. Joshi, 1992: A formal Look at Dependency Grammars and Phrase-Structure Grammars, with Special Consideration of Word- Order Phenomena. in Leo Wanner (ed.) Rece111 Trends i11 Meaning-Text Theory", | |
| "links": null | |
| }, | |
| "BIBREF11": { | |
| "ref_id": "b11", | |
| "title": "re, 1959: Elements de sy11taxe structurale", | |
| "authors": [ | |
| { | |
| "first": "L", | |
| "middle": [], | |
| "last": "", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1995, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "L. Tesnil:re, 1959: Elements de sy11taxe structurale, Klincksieck, Paris. (RVW95) 0. Rambow, K. Vijay-Shanker. D. Wcir. 1995: D-Tree Grammars, ACL'95.", | |
| "links": null | |
| }, | |
| "BIBREF12": { | |
| "ref_id": "b12", | |
| "title": "An alternative conception of tree-adjoining derivation", | |
| "authors": [ | |
| { | |
| "first": "S", | |
| "middle": [], | |
| "last": "Shieber", | |
| "suffix": "" | |
| }, | |
| { | |
| "first": "Y", | |
| "middle": [], | |
| "last": "Schabes", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": 1994, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "S. Shieber, Y. Schabes, 1994: An alternative conception of tree-adjoining derivation.", | |
| "links": null | |
| }, | |
| "BIBREF13": { | |
| "ref_id": "b13", | |
| "title": "Computational Li11g11istics, 20. 1. (XTAG95) XTAG research group: A Lexicalizcd TAG for", | |
| "authors": [], | |
| "year": null, | |
| "venue": "", | |
| "volume": "", | |
| "issue": "", | |
| "pages": "", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "Computational Li11g11istics, 20. 1. (XTAG95) XTAG research group: A Lexicalizcd TAG for [nglish. Technical report IRCS 95-03. \u2022 Univ. of Pennsylvania. (Online updntcd vcrsion).", | |
| "links": null | |
| }, | |
| "BIBREF14": { | |
| "ref_id": "b14", | |
| "title": "1967 : 0 semanticeskom sinteze [On semantic synthesis J. Problemy kybemetiki, v", | |
| "authors": [ | |
| { | |
| "first": "A", | |
| "middle": [], | |
| "last": "Zolkovskij", | |
| "suffix": "" | |
| } | |
| ], | |
| "year": null, | |
| "venue": "", | |
| "volume": "19", | |
| "issue": "", | |
| "pages": "177--238", | |
| "other_ids": {}, | |
| "num": null, | |
| "urls": [], | |
| "raw_text": "A. Zolkovskij, L Mel'cuk. 1967 : 0 semanticeskom sinteze [On semantic synthesis J. Problemy kybemetiki, v. 19, 177-238.", | |
| "links": null | |
| } | |
| }, | |
| "ref_entries": { | |
| "FIGREF0": { | |
| "uris": null, | |
| "type_str": "figure", | |
| "num": null, | |
| "text": "Figure 2 : DSyntS" | |
| }, | |
| "FIGREF2": { | |
| "uris": null, | |
| "type_str": "figure", | |
| "num": null, | |
| "text": "DT (left) and SemS (right),with a different lay out to facilitate comparison" | |
| }, | |
| "FIGREF3": { | |
| "uris": null, | |
| "type_str": "figure", | |
| "num": null, | |
| "text": "Sems of (4)" | |
| }, | |
| "TABREF1": { | |
| "html": null, | |
| "num": null, | |
| "type_str": "table", | |
| "text": "dependency is valid only if adjunction occurs on the spine of the tree receiving adjunction. This is the case most of the time. Yct wc thank Martine Smets for pointing to us a problcmmic casc:in Paul givesflowers only to Mary. to is scmantirnlly cmpty and appears as co-head in thc gi1\u2022e lrcc. Thc adverb 011/y adjoins on the PP node of thc gi1\u2022e trcc though it scmantically governs Mary.", | |
| "content": "<table><tr><td/><td/><td/><td/><td colspan=\"2\">the DT and SemS f\u00fcr</td></tr><tr><td colspan=\"2\">sentence ( l ). 9</td><td/><td/><td/></tr><tr><td>subslitution site tree</td><td/><td colspan=\"2\">semamic-govemor</td><td>adjunclion site cn:e</td><td>semanlic de~ndent</td></tr><tr><td>1</td><td/><td/><td/><td/></tr><tr><td>1 1 1 substituted</td><td>=</td><td colspan=\"2\">J semanlic</td><td>adjoined</td><td>= 1 semantic</td></tr><tr><td>tree</td><td colspan=\"3\">deoendent</td><td>tn:.:</td><td>gov.:mor</td></tr><tr><td/><td/><td/><td/><td/><td>), where</td></tr><tr><td colspan=\"6\">multiple modifier adjunctions are allowcd at the</td></tr><tr><td colspan=\"2\">same address. 10</td><td/><td/><td/></tr><tr><td colspan=\"6\">1 The fact that the DT should represent semantics is</td></tr><tr><td colspan=\"6\">not new. See for example (A93) who distinguishes</td></tr><tr><td colspan=\"6\">between g/ass in a wine glass anu in a glass of wine on</td></tr><tr><td colspan=\"6\">purely semantic grounds; (K89) who mentions <hat</td></tr><tr><td colspan=\"6\">TAG should \"preserve a straightforward compositional</td></tr><tr><td colspan=\"5\">semantics\"; (098) who descirbes</td><td>G-TAG. a</td></tr><tr><td colspan=\"6\">generation system based on TAG where a derivation</td></tr><tr><td colspan=\"6\">tree is built by lexicalizing a conceptual structurc.</td></tr><tr><td colspan=\"6\">9 The TAG analysis is from (X95), except that</td></tr><tr><td colspan=\"3\">detenniners are</td><td colspan=\"2\">not considered</td><td>as</td><td>nominal</td></tr><tr><td colspan=\"5\">complements and are thus adjoined.</td></tr></table>" | |
| } | |
| } | |
| } | |
| } |