{"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/r2-d2-a-modular-baseline-for-open-domain","title":"R2-D2: A Modular Baseline for Open-Domain Question Answering","arxiv_id":"2109.03502","date":"2021-09-08","proceeding":"Findings (EMNLP) 2021 11","authors":["Martin Fajcik","Martin Docekal","Karel Ondrej","Pavel Smrz"],"abstract":"This work presents a novel four-stage open-domain QA pipeline R2-D2 (Rank twice, reaD twice). The pipeline is composed of a retriever, passage reranker, extractive reader, generative reader and a mechanism that aggregates the final prediction from all system's components. We demonstrate its strength across three open-domain QA datasets: NaturalQuestions, TriviaQA and EfficientQA, surpassing state-of-the-art on the first two. Our analysis demonstrates that: (i) combining extractive and generative reader yields absolute improvements up to 5 exact match and it is at least twice as effective as the posterior averaging ensemble of the same models with different parameters, (ii) the extractive reader with fewer parameters can match the performance of the generative reader on extractive QA datasets.","url_abs":"https://arxiv.org/abs/2109.03502v1","url_pdf":"https://arxiv.org/pdf/2109.03502v1.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":"r2-d2-a-modular-baseline-for-open-domain","repo_url":"https://github.com/KNOT-FIT-BUT/R2-D2","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"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":"triviaqa","task_name":"TriviaQA"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/open-domain-question-answering-on-natural","task":"Open-Domain Question Answering","dataset":"Natural Questions","model":"R2-D2 \\w HN-DPR","rank_in_archive_order":2,"of":5,"metrics":{"Exact Match":"55.9"},"uses_additional_data":false},{"leaderboard":"/sota/passage-retrieval-on-natural-questions","task":"Passage Retrieval","dataset":"Natural Questions","model":"DPR+ELECTRA-large-extreader-reranker","rank_in_archive_order":5,"of":10,"metrics":{"Precision@100":"88.25","Precision@20":"85.26"},"uses_additional_data":false},{"leaderboard":"/sota/passage-retrieval-on-natural-questions","task":"Passage Retrieval","dataset":"Natural Questions","model":"DPR+RoBERTa-base-crossencoder-reranker","rank_in_archive_order":6,"of":10,"metrics":{"Precision@100":"88.03","Precision@20":"84.46"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-natural-questions","task":"Question Answering","dataset":"Natural Questions","model":"R2-D2 (full)","rank_in_archive_order":5,"of":47,"metrics":{"EM":"55.9"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-natural-questions-long","task":"Question Answering","dataset":"Natural Questions (long)","model":"R2-D2 \\w HN-DPR","rank_in_archive_order":10,"of":13,"metrics":{"EM":"55.9"},"uses_additional_data":false}],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=2109.03502","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}