{"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/preference-leakage-a-contamination-problem-in","title":"Preference Leakage: A Contamination Problem in LLM-as-a-judge","arxiv_id":"2502.01534","date":"2025-02-03","proceeding":null,"authors":["Dawei Li","Renliang Sun","Yue Huang","Ming Zhong","Bohan Jiang","Jiawei Han","Xiangliang Zhang","Wei Wang","Huan Liu"],"abstract":"Large Language Models (LLMs) as judges and LLM-based data synthesis have emerged as two fundamental LLM-driven data annotation methods in model development. While their combination significantly enhances the efficiency of model training and evaluation, little attention has been given to the potential contamination brought by this new model development paradigm. In this work, we expose preference leakage, a contamination problem in LLM-as-a-judge caused by the relatedness between the synthetic data generators and LLM-based evaluators. To study this issue, we first define three common relatednesses between data generator LLM and judge LLM: being the same model, having an inheritance relationship, and belonging to the same model family. Through extensive experiments, we empirically confirm the bias of judges towards their related student models caused by preference leakage across multiple LLM baselines and benchmarks. Further analysis suggests that preference leakage is a pervasive issue that is harder to detect compared to previously identified biases in LLM-as-a-judge scenarios. All of these findings imply that preference leakage is a widespread and challenging problem in the area of LLM-as-a-judge. We release all codes and data at: https://github.com/David-Li0406/Preference-Leakage.","url_abs":"https://arxiv.org/abs/2502.01534v1","url_pdf":"https://arxiv.org/pdf/2502.01534v1.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":"preference-leakage-a-contamination-problem-in","repo_url":"https://github.com/david-li0406/preference-leakage","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[],"methods":[{"method_slug":"attention","method_name":"Attention"},{"method_slug":"softmax","method_name":"Softmax"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2502.01534","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2502.01534"}},"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/David-Li0406/Preference-Leakage","reach":{"status":"ok"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/david-li0406/preference-leakage","reach":{"status":"ok"}}],"summary":{"ran_draft_wrong":1,"unverified":2},"by_repo_kind":{"official":{"samples":3,"ran":1,"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":"0c77763ca4cd8bc4","entry":"get_metrics_evaluator","repo":"david-li0406/preference-leakage","repo_kind":"official","path":"evaluation/Alpaca_Eval2/src/alpaca_eval/analyze.py","file_url":"https://github.com/david-li0406/preference-leakage/blob/HEAD/evaluation/Alpaca_Eval2/src/alpaca_eval/analyze.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":"0c77763ca4cd8bc4"}},{"code_sha256_prefix":"0eb53db38f0faafd","entry":"get_annotations","repo":"david-li0406/preference-leakage","repo_kind":"official","path":"evaluation/Alpaca_Eval2/src/alpaca_eval/analyze.py","file_url":"https://github.com/david-li0406/preference-leakage/blob/HEAD/evaluation/Alpaca_Eval2/src/alpaca_eval/analyze.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":"0eb53db38f0faafd"}},{"code_sha256_prefix":"de8a267bb7f7d4bd","entry":"get_crossannotations","repo":"david-li0406/preference-leakage","repo_kind":"official","path":"evaluation/Alpaca_Eval2/src/alpaca_eval/analyze.py","file_url":"https://github.com/david-li0406/preference-leakage/blob/HEAD/evaluation/Alpaca_Eval2/src/alpaca_eval/analyze.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":"de8a267bb7f7d4bd"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}