{"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/robust-pronoun-use-fidelity-with-english-llms","title":"Robust Pronoun Fidelity with English LLMs: Are they Reasoning, Repeating, or Just Biased?","arxiv_id":"2404.03134","date":"2024-04-04","proceeding":null,"authors":["Vagrant Gautam","Eileen Bingert","Dawei Zhu","Anne Lauscher","Dietrich Klakow"],"abstract":"Robust, faithful and harm-free pronoun use for individuals is an important goal for language model development as their use increases, but prior work tends to study only one or two of these characteristics at a time. To measure progress towards the combined goal, we introduce the task of pronoun fidelity: given a context introducing a co-referring entity and pronoun, the task is to reuse the correct pronoun later. We present RUFF, a carefully-designed dataset of over 5 million instances to measure robust pronoun fidelity in English, and we evaluate 37 model variants from nine popular families, across architectures (encoder-only, decoder-only and encoder-decoder) and scales (11M-70B parameters). When an individual is introduced with a pronoun, models can mostly faithfully reuse this pronoun in the next sentence, but they are significantly worse with she/her/her, singular they and neopronouns. Moreover, models are easily distracted by non-adversarial sentences discussing other people; even one sentence with a distractor pronoun causes accuracy to drop on average by 34 percentage points. Our results show that pronoun fidelity is not robust, in a simple, naturalistic setting where humans achieve nearly 100% accuracy. We encourage researchers to bridge the gaps we find and to carefully evaluate reasoning in settings where superficial repetition might inflate perceptions of model performance.","url_abs":"https://arxiv.org/abs/2404.03134v3","url_pdf":"https://arxiv.org/pdf/2404.03134v3.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":"robust-pronoun-use-fidelity-with-english-llms","repo_url":"https://github.com/uds-lsv/pronoun-use-fidelity","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"AGPL-3.0"}},{"paper_slug":"robust-pronoun-use-fidelity-with-english-llms","repo_url":"https://github.com/uds-lsv/robust-pronoun-fidelity","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[{"slug":"ruff","name":"RUFF","full_name":""}],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2404.03134","atlas_url":"https://app.syntology.ai/?focus=2404.03134","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2404.03134"}},"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/uds-lsv/pronoun-use-fidelity","reach":{"status":"ok","spdx":"AGPL-3.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/uds-lsv/robust-pronoun-fidelity","reach":null}],"summary":{"ran_draft_wrong":3},"by_repo_kind":{"official":{"samples":3,"ran":3,"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":3,"samples":[{"code_sha256_prefix":"01f7934e47e548f1","entry":"get_mask_token","repo":"uds-lsv/robust-pronoun-fidelity","repo_kind":"official","path":"scripts/score_models.py","file_url":"https://github.com/uds-lsv/robust-pronoun-fidelity/blob/HEAD/scripts/score_models.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":"AGPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"01f7934e47e548f1"}},{"code_sha256_prefix":"08ce9f44fb0f186c","entry":"get_output_line","repo":"uds-lsv/robust-pronoun-fidelity","repo_kind":"official","path":"scripts/add_context.py","file_url":"https://github.com/uds-lsv/robust-pronoun-fidelity/blob/HEAD/scripts/add_context.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":"AGPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"08ce9f44fb0f186c"}},{"code_sha256_prefix":"8a2b4155f22d86aa","entry":"instantiate_template","repo":"uds-lsv/robust-pronoun-fidelity","repo_kind":"official","path":"scripts/add_context.py","file_url":"https://github.com/uds-lsv/robust-pronoun-fidelity/blob/HEAD/scripts/add_context.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"AGPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"8a2b4155f22d86aa"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}