{"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/mitigating-the-bias-of-large-language-model","title":"Mitigating the Bias of Large Language Model Evaluation","arxiv_id":"2409.16788","date":"2024-09-25","proceeding":null,"authors":["Hongli Zhou","Hui Huang","Yunfei Long","Bing Xu","Conghui Zhu","Hailong Cao","Muyun Yang","Tiejun Zhao"],"abstract":"Recently, there has been a trend of evaluating the Large Language Model (LLM) quality in the flavor of LLM-as-a-Judge, namely leveraging another LLM to evaluate the current output quality. However, existing judges are proven to be biased, namely they would favor answers which present better superficial quality (such as verbosity, fluency) while ignoring the instruction following ability. In this work, we propose systematic research about the bias of LLM-as-a-Judge. Specifically, for closed-source judge models, we apply calibration to mitigate the significance of superficial quality, both on probability level and prompt level. For open-source judge models, we propose to mitigate the bias by contrastive training, with curated negative samples that deviate from instruction but present better superficial quality. We apply our methods on the bias evaluation benchmark, and experiment results show our methods mitigate the bias by a large margin while maintaining a satisfactory evaluation accuracy.","url_abs":"https://arxiv.org/abs/2409.16788v1","url_pdf":"https://arxiv.org/pdf/2409.16788v1.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":"mitigating-the-bias-of-large-language-model","repo_url":"https://github.com/Joe-Hall-Lee/Debias","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"instruction-following","task_name":"Instruction Following"},{"task_slug":"language-model-evaluation","task_name":"Language Model Evaluation"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"large-language-model","task_name":"Large Language Model"},{"task_slug":"model","task_name":"model"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2409.16788","atlas_url":"https://app.syntology.ai/?focus=2409.16788","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2409.16788"}},"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/Joe-Hall-Lee/Debias","reach":{"status":"ok"}}],"summary":{"ran":1,"ran_honours":1},"by_repo_kind":{"official":{"samples":2,"ran":2,"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":2,"samples":[{"code_sha256_prefix":"860708bdff892833","entry":"compute_difference","repo":"Joe-Hall-Lee/Debias","repo_kind":"official","path":"Calibration/reverse.py","file_url":"https://github.com/Joe-Hall-Lee/Debias/blob/HEAD/Calibration/reverse.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":"860708bdff892833"}},{"code_sha256_prefix":"b904d307320eb547","entry":"filt_pred_score_list_by_gt_score_list","repo":"Joe-Hall-Lee/Debias","repo_kind":"official","path":"Calibration/Superficial_CoT/eval_model_judgement.py","file_url":"https://github.com/Joe-Hall-Lee/Debias/blob/HEAD/Calibration/Superficial_CoT/eval_model_judgement.py","link_basis":"plan_row","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"b904d307320eb547"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}