{"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/hellaswag-can-a-machine-really-finish-your","title":"HellaSwag: Can a Machine Really Finish Your Sentence?","arxiv_id":"1905.07830","date":"2019-05-19","proceeding":"ACL 2019 7","authors":["Rowan Zellers","Ari Holtzman","Yonatan Bisk","Ali Farhadi","Yejin Choi"],"abstract":"Recent work by Zellers et al. (2018) introduced a new task of commonsense natural language inference: given an event description such as \"A woman sits at a piano,\" a machine must select the most likely followup: \"She sets her fingers on the keys.\" With the introduction of BERT, near human-level performance was reached. Does this mean that machines can perform human level commonsense inference? In this paper, we show that commonsense inference still proves difficult for even state-of-the-art models, by presenting HellaSwag, a new challenge dataset. Though its questions are trivial for humans (>95% accuracy), state-of-the-art models struggle (<48%). We achieve this via Adversarial Filtering (AF), a data collection paradigm wherein a series of discriminators iteratively select an adversarial set of machine-generated wrong answers. AF proves to be surprisingly robust. The key insight is to scale up the length and complexity of the dataset examples towards a critical 'Goldilocks' zone wherein generated text is ridiculous to humans, yet often misclassified by state-of-the-art models. Our construction of HellaSwag, and its resulting difficulty, sheds light on the inner workings of deep pretrained models. More broadly, it suggests a new path forward for NLP research, in which benchmarks co-evolve with the evolving state-of-the-art in an adversarial way, so as to present ever-harder challenges.","url_abs":"https://arxiv.org/abs/1905.07830v1","url_pdf":"https://arxiv.org/pdf/1905.07830v1.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":"hellaswag-can-a-machine-really-finish-your","repo_url":"https://github.com/PlusLabNLP/Plot-guided-Coherence-Evaluation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"hellaswag-can-a-machine-really-finish-your","repo_url":"https://github.com/facebookresearch/text_characterization_toolkit","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"hellaswag","task_name":"HellaSwag"},{"task_slug":"natural-language-inference","task_name":"Natural Language Inference"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"sentence-completion","task_name":"Sentence Completion"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bert","method_name":"BERT"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-linear-decay","method_name":"Linear Warmup With Linear Decay"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"weight-decay","method_name":"Weight Decay"},{"method_slug":"wordpiece","method_name":"WordPiece"}],"datasets_introduced":[{"slug":"hellaswag","name":"HellaSwag","full_name":""}],"methods_introduced":[],"results":[{"leaderboard":"/sota/sentence-completion-on-hellaswag","task":"Sentence Completion","dataset":"HellaSwag","model":"BERT-Large 340M","rank_in_archive_order":71,"of":89,"metrics":{"Accuracy":"47.3"},"uses_additional_data":false},{"leaderboard":"/sota/sentence-completion-on-hellaswag","task":"Sentence Completion","dataset":"HellaSwag","model":"GPT-1 117M","rank_in_archive_order":73,"of":89,"metrics":{"Accuracy":"41.7"},"uses_additional_data":false},{"leaderboard":"/sota/sentence-completion-on-hellaswag","task":"Sentence Completion","dataset":"HellaSwag","model":"BERT-Base 110M","rank_in_archive_order":77,"of":89,"metrics":{"Accuracy":"40.5"},"uses_additional_data":false},{"leaderboard":"/sota/sentence-completion-on-hellaswag","task":"Sentence Completion","dataset":"HellaSwag","model":"LSTM + BERT-Base","rank_in_archive_order":80,"of":89,"metrics":{"Accuracy":"36.2"},"uses_additional_data":false},{"leaderboard":"/sota/sentence-completion-on-hellaswag","task":"Sentence Completion","dataset":"HellaSwag","model":"ESIM + ElMo","rank_in_archive_order":82,"of":89,"metrics":{"Accuracy":"33.3"},"uses_additional_data":false},{"leaderboard":"/sota/sentence-completion-on-hellaswag","task":"Sentence Completion","dataset":"HellaSwag","model":"LSTM + GloVe","rank_in_archive_order":84,"of":89,"metrics":{"Accuracy":"31.7"},"uses_additional_data":false},{"leaderboard":"/sota/sentence-completion-on-hellaswag","task":"Sentence Completion","dataset":"HellaSwag","model":"fastText","rank_in_archive_order":85,"of":89,"metrics":{"Accuracy":"31.6"},"uses_additional_data":false},{"leaderboard":"/sota/sentence-completion-on-hellaswag","task":"Sentence Completion","dataset":"HellaSwag","model":"LSTM + ElMo","rank_in_archive_order":86,"of":89,"metrics":{"Accuracy":"31.4"},"uses_additional_data":false},{"leaderboard":"/sota/sentence-completion-on-hellaswag","task":"Sentence Completion","dataset":"HellaSwag","model":"Random chance baseline","rank_in_archive_order":89,"of":89,"metrics":{"Accuracy":"25"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1905.07830","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1905.07830"}},"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":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/facebookresearch/text_characterization_toolkit","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/PlusLabNLP/Plot-guided-Coherence-Evaluation","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":6},"by_repo_kind":{"listed":{"samples":6,"ran":0,"repositories":2}},"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":4,"samples":[{"code_sha256_prefix":"ca9c20fdb4d8828a","entry":"evaluate","repo":"PlusLabNLP/Plot-guided-Coherence-Evaluation","repo_kind":"listed","path":"run_glue.py","file_url":"https://github.com/PlusLabNLP/Plot-guided-Coherence-Evaluation/blob/HEAD/run_glue.py","link_basis":"harvester_set","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":"ca9c20fdb4d8828a"}},{"code_sha256_prefix":"bbe650a2d1efc562","entry":"mean_or_none","repo":"facebookresearch/text_characterization_toolkit","repo_kind":"listed","path":"text_characterization/utils.py","file_url":"https://github.com/facebookresearch/text_characterization_toolkit/blob/HEAD/text_characterization/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":false,"mcp_get_code":{"code_sha256":"bbe650a2d1efc562"}},{"code_sha256_prefix":"0bb699cebf314bf6","entry":"plot_data","repo":"facebookresearch/text_characterization_toolkit","repo_kind":"listed","path":"text_characterization/analysis.py","file_url":"https://github.com/facebookresearch/text_characterization_toolkit/blob/HEAD/text_characterization/analysis.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":false,"mcp_get_code":{"code_sha256":"0bb699cebf314bf6"}},{"code_sha256_prefix":"2fe1321cdb7f774d","entry":"safe_div","repo":"facebookresearch/text_characterization_toolkit","repo_kind":"listed","path":"text_characterization/utils.py","file_url":"https://github.com/facebookresearch/text_characterization_toolkit/blob/HEAD/text_characterization/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":false,"mcp_get_code":{"code_sha256":"2fe1321cdb7f774d"}},{"code_sha256_prefix":"fb6fdbc437bec260","entry":"safe_stdev","repo":"facebookresearch/text_characterization_toolkit","repo_kind":"listed","path":"text_characterization/utils.py","file_url":"https://github.com/facebookresearch/text_characterization_toolkit/blob/HEAD/text_characterization/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":false,"mcp_get_code":{"code_sha256":"fb6fdbc437bec260"}},{"code_sha256_prefix":"b0538d4790ceabf1","entry":"test_text_rel_score","repo":"PlusLabNLP/Plot-guided-Coherence-Evaluation","repo_kind":"listed","path":"run_glue.py","file_url":"https://github.com/PlusLabNLP/Plot-guided-Coherence-Evaluation/blob/HEAD/run_glue.py","link_basis":"harvester_set","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":"b0538d4790ceabf1"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}