{"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/offsetbias-leveraging-debiased-data-for","title":"OffsetBias: Leveraging Debiased Data for Tuning Evaluators","arxiv_id":"2407.06551","date":"2024-07-09","proceeding":null,"authors":["Junsoo Park","Seungyeon Jwa","Meiying Ren","Daeyoung Kim","Sanghyuk Choi"],"abstract":"Employing Large Language Models (LLMs) to assess the quality of generated responses, such as prompting instruct-tuned models or fine-tuning judge models, has become a widely adopted evaluation method. It is also known that such evaluators are vulnerable to biases, such as favoring longer responses. While it is important to overcome this problem, the specifics of these biases remain under-explored. In this work, we qualitatively identify six types of biases inherent in various judge models. We propose EvalBiasBench as a meta-evaluation collection of hand-crafted test cases for each bias type. Additionally, we present de-biasing dataset construction methods and the associated preference dataset OffsetBias. Experimental results demonstrate that fine-tuning on our dataset significantly enhances the robustness of judge models against biases and improves performance across most evaluation scenarios. We release our datasets and the fine-tuned judge model to public.","url_abs":"https://arxiv.org/abs/2407.06551v2","url_pdf":"https://arxiv.org/pdf/2407.06551v2.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":"offsetbias-leveraging-debiased-data-for","repo_url":"https://github.com/ncsoft/offsetbias","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"paper_slug":"offsetbias-leveraging-debiased-data-for","repo_url":"https://github.com/RLHFlow/RLHF-Reward-Modeling","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"offsetbias-leveraging-debiased-data-for","repo_url":"https://github.com/atla-ai/selene-mini","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2407.06551","atlas_url":"https://app.syntology.ai/?focus=2407.06551","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2407.06551"}},"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/RLHFlow/RLHF-Reward-Modeling","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ncsoft/offsetbias","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/atla-ai/selene-mini","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"ran":2,"unverified":1},"by_repo_kind":{"official":{"samples":3,"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":0,"samples":[{"code_sha256_prefix":"34978fa5a31a0e95","entry":"get_model_statistics","repo":"ncsoft/offsetbias","repo_kind":"official","path":"run_bench.py","file_url":"https://github.com/ncsoft/offsetbias/blob/HEAD/run_bench.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"34978fa5a31a0e95"}},{"code_sha256_prefix":"2c4b0375a2ce745c","entry":"make_data_row","repo":"ncsoft/offsetbias","repo_kind":"official","path":"run_bench.py","file_url":"https://github.com/ncsoft/offsetbias/blob/HEAD/run_bench.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"2c4b0375a2ce745c"}},{"code_sha256_prefix":"54252129bd845012","entry":"get_benchmark_data","repo":"ncsoft/offsetbias","repo_kind":"official","path":"run_bench.py","file_url":"https://github.com/ncsoft/offsetbias/blob/HEAD/run_bench.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"54252129bd845012"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}