{"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/unqovering-stereotyping-biases-via","title":"UnQovering Stereotyping Biases via Underspecified Questions","arxiv_id":"2010.02428","date":"2020-10-06","proceeding":"Findings of the Association for Computational Linguistics 2020","authors":["Tao Li","Tushar Khot","Daniel Khashabi","Ashish Sabharwal","Vivek Srikumar"],"abstract":"While language embeddings have been shown to have stereotyping biases, how these biases affect downstream question answering (QA) models remains unexplored. We present UNQOVER, a general framework to probe and quantify biases through underspecified questions. We show that a naive use of model scores can lead to incorrect bias estimates due to two forms of reasoning errors: positional dependence and question independence. We design a formalism that isolates the aforementioned errors. As case studies, we use this metric to analyze four important classes of stereotypes: gender, nationality, ethnicity, and religion. We probe five transformer-based QA models trained on two QA datasets, along with their underlying language models. Our broad study reveals that (1) all these models, with and without fine-tuning, have notable stereotyping biases in these classes; (2) larger models often have higher bias; and (3) the effect of fine-tuning on bias varies strongly with the dataset and the model size.","url_abs":"https://arxiv.org/abs/2010.02428v3","url_pdf":"https://arxiv.org/pdf/2010.02428v3.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":"unqovering-stereotyping-biases-via","repo_url":"https://github.com/allenai/unqover","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"question-answering","task_name":"Question Answering"}],"methods":[],"datasets_introduced":[{"slug":"unqover","name":"UnQover","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2010.02428","atlas_url":"https://app.syntology.ai/?focus=2010.02428","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2010.02428"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/allenai/unqover","reach":null}],"summary":{"ran_draft_wrong":4},"by_repo_kind":{"official":{"samples":4,"ran":4,"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":0,"samples":[{"code_sha256_prefix":"cf56d483faf1a2b7","entry":"compose_subj_act_bij_line","repo":"allenai/unqover","repo_kind":"official","path":"templates/generate_underspecified_templates.py","file_url":"https://github.com/allenai/unqover/blob/HEAD/templates/generate_underspecified_templates.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"cf56d483faf1a2b7"}},{"code_sha256_prefix":"293269302d7c4974","entry":"load_gender_names","repo":"allenai/unqover","repo_kind":"official","path":"lm/predict.py","file_url":"https://github.com/allenai/unqover/blob/HEAD/lm/predict.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"293269302d7c4974"}},{"code_sha256_prefix":"3cc773e67543344d","entry":"load_input","repo":"allenai/unqover","repo_kind":"official","path":"lm/predict.py","file_url":"https://github.com/allenai/unqover/blob/HEAD/lm/predict.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3cc773e67543344d"}},{"code_sha256_prefix":"4018efb99d4c77a3","entry":"preprocess","repo":"allenai/unqover","repo_kind":"official","path":"lm/predict.py","file_url":"https://github.com/allenai/unqover/blob/HEAD/lm/predict.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4018efb99d4c77a3"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}