{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/leveraging-passage-retrieval-with-generative","title":"Leveraging Passage Retrieval with Generative Models for Open Domain Question Answering","arxiv_id":"2007.01282","date":"2020-07-02","proceeding":"EACL 2021 2","authors":["Gautier Izacard","Edouard Grave"],"abstract":"Generative models for open domain question answering have proven to be competitive, without resorting to external knowledge. While promising, this approach requires to use models with billions of parameters, which are expensive to train and query. In this paper, we investigate how much these models can benefit from retrieving text passages, potentially containing evidence. We obtain state-of-the-art results on the Natural Questions and TriviaQA open benchmarks. Interestingly, we observe that the performance of this method significantly improves when increasing the number of retrieved passages. This is evidence that generative models are good at aggregating and combining evidence from multiple passages.","url_abs":"https://arxiv.org/abs/2007.01282v2","url_pdf":"https://arxiv.org/pdf/2007.01282v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"leveraging-passage-retrieval-with-generative","repo_url":"https://github.com/FenQQQ/Fusion-in-decoder","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"leveraging-passage-retrieval-with-generative","repo_url":"https://github.com/ZIZUN/MAFiD","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"leveraging-passage-retrieval-with-generative","repo_url":"https://github.com/amzn/refuel-open-domain-qa","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"leveraging-passage-retrieval-with-generative","repo_url":"https://github.com/facebookresearch/FiD","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"NOASSERTION"}},{"paper_slug":"leveraging-passage-retrieval-with-generative","repo_url":"https://github.com/jhyuklee/DensePhrases","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"leveraging-passage-retrieval-with-generative","repo_url":"https://github.com/princeton-nlp/DensePhrases","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"leveraging-passage-retrieval-with-generative","repo_url":"https://github.com/uclnlp/APE","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"leveraging-passage-retrieval-with-generative","repo_url":"https://github.com/xfactlab/emnlp2023-damaging-retrieval","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"natural-questions","task_name":"Natural Questions"},{"task_slug":"open-domain-question-answering","task_name":"Open-Domain Question Answering"},{"task_slug":"passage-retrieval","task_name":"Passage Retrieval"},{"task_slug":"question-answering","task_name":"Question Answering"},{"task_slug":"retrieval","task_name":"Retrieval"},{"task_slug":"triviaqa","task_name":"TriviaQA"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/question-answering-on-conditionalqa","task":"Question Answering","dataset":"ConditionalQA","model":"FiD","rank_in_archive_order":1,"of":3,"metrics":{"Conditional (answers)":"45.2 / 49.7","Conditional (w/ conditions)":"4.7 / 5.8","Overall (answers)":"44.4 / 50.8","Overall (w/ conditions)":"35.0 / 40.6"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-natural-questions","task":"Question Answering","dataset":"Natural Questions","model":"FiD-KD (full)","rank_in_archive_order":7,"of":47,"metrics":{"EM":"54.7"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-natural-questions","task":"Question Answering","dataset":"Natural Questions","model":"FID (full)","rank_in_archive_order":10,"of":47,"metrics":{"EM":"51.4"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-triviaqa","task":"Question Answering","dataset":"TriviaQA","model":"Fusion-in-Decoder (large)","rank_in_archive_order":35,"of":56,"metrics":{"EM":"67.6"},"uses_additional_data":false}],"syntology":{"syntology_url":"https://syntology.ai/paper/2007.01282","atlas_url":"https://app.syntology.ai/?focus=2007.01282","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}