{"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/unifiedqa-crossing-format-boundaries-with-a","title":"UnifiedQA: Crossing Format Boundaries With a Single QA System","arxiv_id":"2005.00700","date":"2020-05-02","proceeding":"Findings of the Association for Computational Linguistics 2020","authors":["Daniel Khashabi","Sewon Min","Tushar Khot","Ashish Sabharwal","Oyvind Tafjord","Peter Clark","Hannaneh Hajishirzi"],"abstract":"Question answering (QA) tasks have been posed using a variety of formats, such as extractive span selection, multiple choice, etc. This has led to format-specialized models, and even to an implicit division in the QA community. We argue that such boundaries are artificial and perhaps unnecessary, given the reasoning abilities we seek to teach are not governed by the format. As evidence, we use the latest advances in language modeling to build a single pre-trained QA model, UnifiedQA, that performs surprisingly well across 17 QA datasets spanning 4 diverse formats. UnifiedQA performs on par with 9 different models that were trained on individual datasets themselves. Even when faced with 12 unseen datasets of observed formats, UnifiedQA performs surprisingly well, showing strong generalization from its out-of-format training data. Finally, simply fine-tuning this pre-trained QA model into specialized models results in a new state of the art on 6 datasets, establishing UnifiedQA as a strong starting point for building QA systems.","url_abs":"https://arxiv.org/abs/2005.00700v3","url_pdf":"https://arxiv.org/pdf/2005.00700v3.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":"unifiedqa-crossing-format-boundaries-with-a","repo_url":"https://github.com/allenai/unifiedqa","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"unifiedqa-crossing-format-boundaries-with-a","repo_url":"https://github.com/facebookresearch/metaicl","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"common-sense-reasoning","task_name":"Common Sense Reasoning"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"multi-task-learning","task_name":"Multi-Task Learning"},{"task_slug":"multi-task-language-understanding","task_name":"Multi-task Language Understanding"},{"task_slug":"multiple-choice","task_name":"Multiple-choice"},{"task_slug":"question-answering","task_name":"Question Answering"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/common-sense-reasoning-on-commonsenseqa","task":"Common Sense Reasoning","dataset":"CommonsenseQA","model":"UnifiedQA 11B (fine-tuned)","rank_in_archive_order":8,"of":38,"metrics":{"Accuracy":"79.1"},"uses_additional_data":true},{"leaderboard":"/sota/common-sense-reasoning-on-commonsenseqa","task":"Common Sense Reasoning","dataset":"CommonsenseQA","model":"T5-XXL 11B (fine-tuned)","rank_in_archive_order":10,"of":38,"metrics":{"Accuracy":"78.1"},"uses_additional_data":false},{"leaderboard":"/sota/common-sense-reasoning-on-commonsenseqa","task":"Common Sense Reasoning","dataset":"CommonsenseQA","model":"UnifiedQA 11B (zero-shot)","rank_in_archive_order":12,"of":38,"metrics":{"Accuracy":"76.2"},"uses_additional_data":false},{"leaderboard":"/sota/common-sense-reasoning-on-commonsenseqa","task":"Common Sense Reasoning","dataset":"CommonsenseQA","model":"UnifiedQA 440M (fine-tuned)","rank_in_archive_order":25,"of":38,"metrics":{"Accuracy":"64"},"uses_additional_data":false},{"leaderboard":"/sota/common-sense-reasoning-on-commonsenseqa","task":"Common Sense Reasoning","dataset":"CommonsenseQA","model":"BART-large 440M (fine-tuned)","rank_in_archive_order":26,"of":38,"metrics":{"Accuracy":"62.5"},"uses_additional_data":false},{"leaderboard":"/sota/common-sense-reasoning-on-winogrande","task":"Common Sense Reasoning","dataset":"WinoGrande","model":"UnifiedQA 11B (fine-tuned)","rank_in_archive_order":5,"of":77,"metrics":{"Accuracy":"89.4"},"uses_additional_data":false},{"leaderboard":"/sota/common-sense-reasoning-on-winogrande","task":"Common Sense Reasoning","dataset":"WinoGrande","model":"Unified QA 406M (fine-tuned)","rank_in_archive_order":32,"of":77,"metrics":{"Accuracy":"73.3"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-openbookqa","task":"Question Answering","dataset":"OpenBookQA","model":"UnifiedQA 11B","rank_in_archive_order":13,"of":45,"metrics":{"Accuracy":"87.2"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-piqa","task":"Question Answering","dataset":"PIQA","model":"UnifiedQA 3B","rank_in_archive_order":11,"of":67,"metrics":{"Accuracy":"85.3"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-social-iqa","task":"Question Answering","dataset":"SIQA","model":"UnifiedQA 3B","rank_in_archive_order":8,"of":24,"metrics":{"Accuracy":"79.8"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2005.00700","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2005.00700"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/allenai/unifiedqa","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/facebookresearch/metaicl","reach":null}],"summary":{"ran":3,"ran_violates":1,"unverified":3},"by_repo_kind":{"official":{"samples":6,"ran":4,"repositories":1},"listed":{"samples":1,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":1,"samples":[{"code_sha256_prefix":"d1731718e8348ee9","entry":"MyDataLoader","repo":"allenai/unifiedqa","repo_kind":"official","path":"bart/unified_data.py","file_url":"https://github.com/allenai/unifiedqa/blob/HEAD/bart/unified_data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"d1731718e8348ee9"}},{"code_sha256_prefix":"3a28a654bb76c5dc","entry":"MyQADataset","repo":"allenai/unifiedqa","repo_kind":"official","path":"bart/unified_data.py","file_url":"https://github.com/allenai/unifiedqa/blob/HEAD/bart/unified_data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3a28a654bb76c5dc"}},{"code_sha256_prefix":"00da23034ccabbe2","entry":"MyUnifiedQADataset","repo":"allenai/unifiedqa","repo_kind":"official","path":"bart/unified_data.py","file_url":"https://github.com/allenai/unifiedqa/blob/HEAD/bart/unified_data.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"00da23034ccabbe2"}},{"code_sha256_prefix":"22b19b62137713f4","entry":"get_exact_match","repo":"allenai/unifiedqa","repo_kind":"official","path":"bart/unified_data.py","file_url":"https://github.com/allenai/unifiedqa/blob/HEAD/bart/unified_data.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"22b19b62137713f4"}},{"code_sha256_prefix":"9ca454d06b418922","entry":"QAData","repo":"allenai/unifiedqa","repo_kind":"official","path":"bart/unified_data.py","file_url":"https://github.com/allenai/unifiedqa/blob/HEAD/bart/unified_data.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9ca454d06b418922"}},{"code_sha256_prefix":"5dd5e28f94edfea8","entry":"UnifiedQAData","repo":"allenai/unifiedqa","repo_kind":"official","path":"bart/unified_data.py","file_url":"https://github.com/allenai/unifiedqa/blob/HEAD/bart/unified_data.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"5dd5e28f94edfea8"}},{"code_sha256_prefix":"327ddf4dd2e27ff5","entry":"process_train","repo":"facebookresearch/metaicl","repo_kind":"listed","path":"preprocess/unifiedqa.py","file_url":"https://github.com/facebookresearch/metaicl/blob/HEAD/preprocess/unifiedqa.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"327ddf4dd2e27ff5"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}