{"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/guardreasoner-towards-reasoning-based-llm","title":"GuardReasoner: Towards Reasoning-based LLM Safeguards","arxiv_id":"2501.18492","date":"2025-01-30","proceeding":null,"authors":["Yue Liu","Hongcheng Gao","Shengfang Zhai","Jun Xia","Tianyi Wu","Zhiwei Xue","Yulin Chen","Kenji Kawaguchi","Jiaheng Zhang","Bryan Hooi"],"abstract":"As LLMs increasingly impact safety-critical applications, ensuring their safety using guardrails remains a key challenge. This paper proposes GuardReasoner, a new safeguard for LLMs, by guiding the guard model to learn to reason. Concretely, we first create the GuardReasonerTrain dataset, which consists of 127K samples with 460K detailed reasoning steps. Then, we introduce reasoning SFT to unlock the reasoning capability of guard models. In addition, we present hard sample DPO to further strengthen their reasoning ability. In this manner, GuardReasoner achieves better performance, explainability, and generalizability. Extensive experiments and analyses on 13 benchmarks of 3 guardrail tasks demonstrate its superiority. Remarkably, GuardReasoner 8B surpasses GPT-4o+CoT by 5.74% and LLaMA Guard 3 8B by 20.84% F1 score on average. We release the training data, code, and models with different scales (1B, 3B, 8B) of GuardReasoner : https://github.com/yueliu1999/GuardReasoner/.","url_abs":"https://arxiv.org/abs/2501.18492v1","url_pdf":"https://arxiv.org/pdf/2501.18492v1.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":"guardreasoner-towards-reasoning-based-llm","repo_url":"https://github.com/yueliu1999/guardreasoner","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[{"method_slug":"dpo","method_name":"DPO"},{"method_slug":"llama","method_name":"LLaMA"},{"method_slug":"sft","method_name":"SFT"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2501.18492","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2501.18492"}},"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/yueliu1999/guardreasoner","reach":null}],"summary":{"ran_honours":1,"ran_draft_wrong":2,"unverified":1},"by_repo_kind":{"official":{"samples":4,"ran":3,"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":4,"samples":[{"code_sha256_prefix":"30338217cdee6677","entry":"cal_weighted_f1","repo":"yueliu1999/guardreasoner","repo_kind":"official","path":"evaluate.py","file_url":"https://github.com/yueliu1999/guardreasoner/blob/HEAD/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"30338217cdee6677"}},{"code_sha256_prefix":"758bc0c482412a97","entry":"post_process","repo":"yueliu1999/guardreasoner","repo_kind":"official","path":"deploy.py","file_url":"https://github.com/yueliu1999/guardreasoner/blob/HEAD/deploy.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"758bc0c482412a97"}},{"code_sha256_prefix":"6f4e6a4cdd7e9015","entry":"reverse","repo":"yueliu1999/guardreasoner","repo_kind":"official","path":"evaluate.py","file_url":"https://github.com/yueliu1999/guardreasoner/blob/HEAD/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6f4e6a4cdd7e9015"}},{"code_sha256_prefix":"f494b9073cfdeb16","entry":"cal_f1","repo":"yueliu1999/guardreasoner","repo_kind":"official","path":"evaluate.py","file_url":"https://github.com/yueliu1999/guardreasoner/blob/HEAD/evaluate.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":"f494b9073cfdeb16"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}