--- tags: - information retrieval - reranking license: apache-2.0 --- # Model Card for TriviaQA Question Encoder in Re2G # Model Details > The approach of RAG, Multi-DPR, and KGI is to train a neural IR (Information Retrieval) component and further train it end-to-end through its impact in generating the correct output. ## Training, Evaluation and Inference The code for training, evaluation and inference is in our github in the [re2g branch](https://github.com/IBM/kgi-slot-filling/tree/re2g). ## Usage The best way to use the model is by adapting the [dpr_apply.py](https://github.com/IBM/kgi-slot-filling/blob/re2g/dpr/dpr_apply.py) ## Citation ``` @inproceedings{glass-etal-2022-re2g, title = "{R}e2{G}: Retrieve, Rerank, Generate", author = "Glass, Michael and Rossiello, Gaetano and Chowdhury, Md Faisal Mahbub and Naik, Ankita and Cai, Pengshan and Gliozzo, Alfio", booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies", month = jul, year = "2022", address = "Seattle, United States", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2022.naacl-main.194", doi = "10.18653/v1/2022.naacl-main.194", pages = "2701--2715", abstract = "As demonstrated by GPT-3 and T5, transformers grow in capability as parameter spaces become larger and larger. However, for tasks that require a large amount of knowledge, non-parametric memory allows models to grow dramatically with a sub-linear increase in computational cost and GPU memory requirements. Recent models such as RAG and REALM have introduced retrieval into conditional generation. These models incorporate neural initial retrieval from a corpus of passages. We build on this line of research, proposing Re2G, which combines both neural initial retrieval and reranking into a BART-based sequence-to-sequence generation. Our reranking approach also permits merging retrieval results from sources with incomparable scores, enabling an ensemble of BM25 and neural initial retrieval. To train our system end-to-end, we introduce a novel variation of knowledge distillation to train the initial retrieval, reranker and generation using only ground truth on the target sequence output. We find large gains in four diverse tasks: zero-shot slot filling, question answering, fact checking and dialog, with relative gains of 9{\%} to 34{\%} over the previous state-of-the-art on the KILT leaderboard. We make our code available as open source.", } ``` ## Model Description The model creators note in the [associated paper](https://aclanthology.org/2022.naacl-main.194.pdf): > As demonstrated by GPT-3 and T5, transformers grow in capability as parameter spaces become larger and larger. However, for tasks that require a large amount of knowledge, non-parametric memory allows models to grow dramatically with a sub-linear increase in computational cost and GPU memory requirements. Recent models such as RAG and REALM have introduced retrieval into conditional generation. These models incorporate neural initial retrieval from a corpus of passages. We build on this line of research, proposing Re2G, which combines both neural initial retrieval and reranking into a BART-based sequence-to-sequence generation. Our reranking approach also permits merging retrieval results from sources with incomparable scores, enabling an ensemble of BM25 and neural initial retrieval. To train our system end-to-end, we introduce a novel variation of knowledge distillation to train the initial retrieval, reranker and generation using only ground truth on the target sequence output. We find large gains in four diverse tasks: zero-shot slot filling, question answering, fact checking and dialog, with relative gains of 9% to 34% over the previous state-of-the-art on the KILT leaderboard. We make our code available as open source. - **Developed by:** IBM - **Shared by [Optional]:** IBM - **Model type:** Query/Passage Reranker - **Language(s) (NLP):** English - **License:** Apache 2.0 - **Parent Model:** [dpr-question_encoder-multiset-base](https://huggingface.co/facebook/dpr-question_encoder-multiset-base) - **Resources for more information:** - [GitHub Repo](https://github.com/IBM/kgi-slot-filling) - [Associated Paper](https://aclanthology.org/2022.naacl-main.194.pdf) # Uses ## Direct Use This model can be used for the task of encoding a question to a vector to be used as a query into an Approximate Nearest Neighbors index. It must be used in combination with a context encoder that encodes passages to a vector and indexes them. # Citation **BibTeX:** ```bibtex @inproceedings{glass-etal-2022-re2g, title = "{R}e2{G}: Retrieve, Rerank, Generate", author = "Glass, Michael and Rossiello, Gaetano and Chowdhury, Md Faisal Mahbub and Naik, Ankita and Cai, Pengshan and Gliozzo, Alfio", booktitle = "Proceedings of the 2022 Conference of the North American Chapter of the Association for Computational Linguistics: Human Language Technologies", month = jul, year = "2022", address = "Seattle, United States", publisher = "Association for Computational Linguistics", url = "https://aclanthology.org/2022.naacl-main.194", doi = "10.18653/v1/2022.naacl-main.194", pages = "2701--2715", abstract = "As demonstrated by GPT-3 and T5, transformers grow in capability as parameter spaces become larger and larger. However, for tasks that require a large amount of knowledge, non-parametric memory allows models to grow dramatically with a sub-linear increase in computational cost and GPU memory requirements. Recent models such as RAG and REALM have introduced retrieval into conditional generation. These models incorporate neural initial retrieval from a corpus of passages. We build on this line of research, proposing Re2G, which combines both neural initial retrieval and reranking into a BART-based sequence-to-sequence generation. Our reranking approach also permits merging retrieval results from sources with incomparable scores, enabling an ensemble of BM25 and neural initial retrieval. To train our system end-to-end, we introduce a novel variation of knowledge distillation to train the initial retrieval, reranker and generation using only ground truth on the target sequence output. We find large gains in four diverse tasks: zero-shot slot filling, question answering, fact checking and dialog, with relative gains of 9{\%} to 34{\%} over the previous state-of-the-art on the KILT leaderboard. We make our code available as open source.", } ```