{"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/jailjudge-a-comprehensive-jailbreak-judge","title":"JAILJUDGE: A Comprehensive Jailbreak Judge Benchmark with Multi-Agent Enhanced Explanation Evaluation Framework","arxiv_id":"2410.12855","date":"2024-10-11","proceeding":null,"authors":["Fan Liu","Yue Feng","Zhao Xu","Lixin Su","Xinyu Ma","Dawei Yin","Hao liu"],"abstract":"Despite advancements in enhancing LLM safety against jailbreak attacks, evaluating LLM defenses remains a challenge, with current methods often lacking explainability and generalization to complex scenarios, leading to incomplete assessments (e.g., direct judgment without reasoning, low F1 score of GPT-4 in complex cases, bias in multilingual scenarios). To address this, we present JAILJUDGE, a comprehensive benchmark featuring diverse risk scenarios, including synthetic, adversarial, in-the-wild, and multilingual prompts, along with high-quality human-annotated datasets. The JAILJUDGE dataset includes over 35k+ instruction-tune data with reasoning explainability and JAILJUDGETEST, a 4.5k+ labeled set for risk scenarios, and a 6k+ multilingual set across ten languages. To enhance evaluation with explicit reasoning, we propose the JailJudge MultiAgent framework, which enables explainable, fine-grained scoring (1 to 10). This framework supports the construction of instruction-tuning ground truth and facilitates the development of JAILJUDGE Guard, an end-to-end judge model that provides reasoning and eliminates API costs. Additionally, we introduce JailBoost, an attacker-agnostic attack enhancer, and GuardShield, a moderation defense, both leveraging JAILJUDGE Guard. Our experiments demonstrate the state-of-the-art performance of JailJudge methods (JailJudge MultiAgent, JAILJUDGE Guard) across diverse models (e.g., GPT-4, Llama-Guard) and zero-shot scenarios. JailBoost and GuardShield significantly improve jailbreak attack and defense tasks under zero-shot settings, with JailBoost enhancing performance by 29.24% and GuardShield reducing defense ASR from 40.46% to 0.15%.","url_abs":"https://arxiv.org/abs/2410.12855v2","url_pdf":"https://arxiv.org/pdf/2410.12855v2.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":"jailjudge-a-comprehensive-jailbreak-judge","repo_url":"https://github.com/usail-hkust/Jailjudge","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"gpt-4","method_name":"GPT-4"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"set","method_name":"SET"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2410.12855","atlas_url":"https://app.syntology.ai/?focus=2410.12855","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2410.12855"}},"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/usail-hkust/Jailjudge","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":9},"by_repo_kind":{"listed":{"samples":9,"ran":9,"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":"ee4b042ddb9385c1","entry":"dempster_combination","repo":"usail-hkust/Jailjudge","repo_kind":"listed","path":"main_multi_agent_judge.py","file_url":"https://github.com/usail-hkust/Jailjudge/blob/HEAD/main_multi_agent_judge.py","link_basis":"harvester_set","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":"ee4b042ddb9385c1"}},{"code_sha256_prefix":"ef7346b6c8e6bc4a","entry":"evident_combination_of_score","repo":"usail-hkust/Jailjudge","repo_kind":"listed","path":"main_multi_agent_judge.py","file_url":"https://github.com/usail-hkust/Jailjudge/blob/HEAD/main_multi_agent_judge.py","link_basis":"harvester_set","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":"ef7346b6c8e6bc4a"}},{"code_sha256_prefix":"75c5d1b87035f6bf","entry":"generate_defense_goal","repo":"usail-hkust/Jailjudge","repo_kind":"listed","path":"defense.py","file_url":"https://github.com/usail-hkust/Jailjudge/blob/HEAD/defense.py","link_basis":"harvester_set","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":"75c5d1b87035f6bf"}},{"code_sha256_prefix":"2a1d9def76e4e4ab","entry":"get_evident_judge_system_prompt","repo":"usail-hkust/Jailjudge","repo_kind":"listed","path":"GPTEvaluator/system_prompts.py","file_url":"https://github.com/usail-hkust/Jailjudge/blob/HEAD/GPTEvaluator/system_prompts.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":"2a1d9def76e4e4ab"}},{"code_sha256_prefix":"4b91f667a68a16ac","entry":"get_init_msg","repo":"usail-hkust/Jailjudge","repo_kind":"listed","path":"GPTEvaluator/common.py","file_url":"https://github.com/usail-hkust/Jailjudge/blob/HEAD/GPTEvaluator/common.py","link_basis":"harvester_set","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":"4b91f667a68a16ac"}},{"code_sha256_prefix":"9fdfb4537f5fd761","entry":"process_target_response","repo":"usail-hkust/Jailjudge","repo_kind":"listed","path":"GPTEvaluator/common.py","file_url":"https://github.com/usail-hkust/Jailjudge/blob/HEAD/GPTEvaluator/common.py","link_basis":"harvester_set","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":"9fdfb4537f5fd761"}},{"code_sha256_prefix":"db66e64c1a095a76","entry":"remove_code_blocks","repo":"usail-hkust/Jailjudge","repo_kind":"listed","path":"models/language_models.py","file_url":"https://github.com/usail-hkust/Jailjudge/blob/HEAD/models/language_models.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":"db66e64c1a095a76"}},{"code_sha256_prefix":"183aa212edc81ba7","entry":"score_to_mass","repo":"usail-hkust/Jailjudge","repo_kind":"listed","path":"main_multi_agent_judge.py","file_url":"https://github.com/usail-hkust/Jailjudge/blob/HEAD/main_multi_agent_judge.py","link_basis":"harvester_set","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":"183aa212edc81ba7"}},{"code_sha256_prefix":"edbf67bdb61110f3","entry":"smooth_pert","repo":"usail-hkust/Jailjudge","repo_kind":"listed","path":"defense.py","file_url":"https://github.com/usail-hkust/Jailjudge/blob/HEAD/defense.py","link_basis":"harvester_set","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":"edbf67bdb61110f3"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}