{"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/bias-augmented-consistency-training-reduces","title":"Bias-Augmented Consistency Training Reduces Biased Reasoning in Chain-of-Thought","arxiv_id":"2403.05518","date":"2024-03-08","proceeding":null,"authors":["James Chua","Edward Rees","Hunar Batra","Samuel R. Bowman","Julian Michael","Ethan Perez","Miles Turpin"],"abstract":"While chain-of-thought prompting (CoT) has the potential to improve the explainability of language model reasoning, it can systematically misrepresent the factors influencing models' behavior--for example, rationalizing answers in line with a user's opinion without mentioning this bias. To mitigate this biased reasoning problem, we introduce bias-augmented consistency training (BCT), an unsupervised fine-tuning scheme that trains models to give consistent reasoning across prompts with and without biasing features. We construct a suite testing nine forms of biased reasoning on seven question-answering tasks, and find that applying BCT to GPT-3.5-Turbo with one bias reduces the rate of biased reasoning by 86% on held-out tasks. Moreover, this model generalizes to other forms of bias, reducing biased reasoning on held-out biases by an average of 37%. As BCT generalizes to held-out biases and does not require gold labels, this method may hold promise for reducing biased reasoning from as-of-yet unknown biases and on tasks where supervision for ground truth reasoning is unavailable.","url_abs":"https://arxiv.org/abs/2403.05518v1","url_pdf":"https://arxiv.org/pdf/2403.05518v1.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":"bias-augmented-consistency-training-reduces","repo_url":"https://github.com/raybears/cot-transparency","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"question-answering","task_name":"Question Answering"}],"methods":[{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"attention-dropout","method_name":"Attention Dropout"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"cosine-annealing","method_name":"Cosine Annealing"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-3","method_name":"GPT-3"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"linear-warmup-with-cosine-annealing","method_name":"Linear Warmup With Cosine Annealing"},{"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"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2403.05518","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2403.05518"}},"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/raybears/cot-transparency","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":6,"unverified":1},"by_repo_kind":{"official":{"samples":7,"ran":6,"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":"027aa06d7b084853","entry":"combine_indicator_with_separator","repo":"raybears/cot-transparency","repo_kind":"official","path":"cot_transparency/data_models/example_base.py","file_url":"https://github.com/raybears/cot-transparency/blob/HEAD/cot_transparency/data_models/example_base.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"027aa06d7b084853"}},{"code_sha256_prefix":"2bd7687eb85b154b","entry":"convert_slist_to_df","repo":"raybears/cot-transparency","repo_kind":"official","path":"cot_transparency/data_models/pd_utils.py","file_url":"https://github.com/raybears/cot-transparency/blob/HEAD/cot_transparency/data_models/pd_utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2bd7687eb85b154b"}},{"code_sha256_prefix":"a2e49f02b84ed98b","entry":"deterministic_hash","repo":"raybears/cot-transparency","repo_kind":"official","path":"cot_transparency/util.py","file_url":"https://github.com/raybears/cot-transparency/blob/HEAD/cot_transparency/util.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"a2e49f02b84ed98b"}},{"code_sha256_prefix":"9cfc88dfbeeda546","entry":"get_exp_dir_name","repo":"raybears/cot-transparency","repo_kind":"official","path":"cot_transparency/util.py","file_url":"https://github.com/raybears/cot-transparency/blob/HEAD/cot_transparency/util.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9cfc88dfbeeda546"}},{"code_sha256_prefix":"b37de94d749b28d8","entry":"is_openai_finetuned","repo":"raybears/cot-transparency","repo_kind":"official","path":"cot_transparency/data_models/config.py","file_url":"https://github.com/raybears/cot-transparency/blob/HEAD/cot_transparency/data_models/config.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"b37de94d749b28d8"}},{"code_sha256_prefix":"3ae749132f5cb915","entry":"setup_logger","repo":"raybears/cot-transparency","repo_kind":"official","path":"cot_transparency/util.py","file_url":"https://github.com/raybears/cot-transparency/blob/HEAD/cot_transparency/util.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"3ae749132f5cb915"}},{"code_sha256_prefix":"814075f8ce28cb2d","entry":"raise_if_not_multiple_choice_answer","repo":"raybears/cot-transparency","repo_kind":"official","path":"cot_transparency/data_models/example_base.py","file_url":"https://github.com/raybears/cot-transparency/blob/HEAD/cot_transparency/data_models/example_base.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"814075f8ce28cb2d"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}