{"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/collapsed-language-models-promote-fairness","title":"Collapsed Language Models Promote Fairness","arxiv_id":"2410.04472","date":"2024-10-06","proceeding":null,"authors":["Jingxuan Xu","Wuyang Chen","Linyi Li","Yao Zhao","Yunchao Wei"],"abstract":"To mitigate societal biases implicitly encoded in recent successful pretrained language models, a diverse array of approaches have been proposed to encourage model fairness, focusing on prompting, data augmentation, regularized fine-tuning, and more. Despite the development, it is nontrivial to reach a principled understanding of fairness and an effective algorithm that can consistently debias language models. In this work, by rigorous evaluations of Neural Collapse -- a learning phenomenon happen in last-layer representations and classifiers in deep networks -- on fairness-related words, we find that debiased language models exhibit collapsed alignment between token representations and word embeddings. More importantly, this observation inspires us to design a principled fine-tuning method that can effectively improve fairness in a wide range of debiasing methods, while still preserving the performance of language models on standard natural language understanding tasks. We attach our code at https://github.com/Xujxyang/Fairness-NC-main.","url_abs":"https://arxiv.org/abs/2410.04472v3","url_pdf":"https://arxiv.org/pdf/2410.04472v3.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":"collapsed-language-models-promote-fairness","repo_url":"https://github.com/Xujxyang/Fairness-NC-main","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"data-augmentation","task_name":"Data Augmentation"},{"task_slug":"fairness","task_name":"Fairness"},{"task_slug":"natural-language-understanding","task_name":"Natural Language Understanding"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2410.04472","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.04472"}},"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/Xujxyang/Fairness-NC-main","reach":{"status":"ok"}}],"summary":{"ran":9,"ran_draft_wrong":3,"unverified":1},"by_repo_kind":{"official":{"samples":13,"ran":12,"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":13,"samples":[{"code_sha256_prefix":"31501071e57d43c4","entry":"calculate_group_to_group_relative_distance_asymmetric","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"adept/distance.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/adept/distance.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"31501071e57d43c4"}},{"code_sha256_prefix":"d6b5586516edd277","entry":"calculate_group_to_group_relative_distance_asymmetric_test","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"adept/distance.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/adept/distance.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d6b5586516edd277"}},{"code_sha256_prefix":"d73ab432982f8191","entry":"calculate_group_to_one_relative_distance_asymmetric_test","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"adept/distance.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/adept/distance.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d73ab432982f8191"}},{"code_sha256_prefix":"5f516b6aaad9acb6","entry":"create_dataset","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"adept/debias.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/adept/debias.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"5f516b6aaad9acb6"}},{"code_sha256_prefix":"f10b5f9242a0af99","entry":"create_nc_dataset","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"adept/debias.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/adept/debias.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f10b5f9242a0af99"}},{"code_sha256_prefix":"f46fa5e6d0948125","entry":"load_and_cache_examples","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"adept/debias_original.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/adept/debias_original.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"f46fa5e6d0948125"}},{"code_sha256_prefix":"fe1027358a8d7950","entry":"process_batch","repo":"xujxyang/fairness-nc-main","repo_kind":"official","path":"bec/code/main_nc.py","file_url":"https://github.com/xujxyang/fairness-nc-main/blob/HEAD/bec/code/main_nc.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"fe1027358a8d7950"}},{"code_sha256_prefix":"d3c361e2a27ed6ed","entry":"process_batch","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"adept/debias.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/adept/debias.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d3c361e2a27ed6ed"}},{"code_sha256_prefix":"8ef531ec328be202","entry":"process_batch","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"adept/lib/collection.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/adept/lib/collection.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"8ef531ec328be202"}},{"code_sha256_prefix":"d0dd7e697cd0cfba","entry":"split_data","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"adept/debias_original.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/adept/debias_original.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d0dd7e697cd0cfba"}},{"code_sha256_prefix":"cc361c2f7b68b11d","entry":"truncate_and_pad","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"adept/lib/collection.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/adept/lib/collection.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"cc361c2f7b68b11d"}},{"code_sha256_prefix":"32967e6aa4441efb","entry":"variable","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"ase/model.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/ase/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"32967e6aa4441efb"}},{"code_sha256_prefix":"42fe7317678380ab","entry":"prepare_tokenizer","repo":"Xujxyang/Fairness-NC-main","repo_kind":"official","path":"adept/collect_sentences.py","file_url":"https://github.com/Xujxyang/Fairness-NC-main/blob/HEAD/adept/collect_sentences.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"42fe7317678380ab"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}