{"id":"jailbreakhunter","title":"JailbreakHunter: A Visual Analytics Approach for Jailbreak Prompts Discovery From Large-Scale Human-LLM Conversational Datasets","authors":[{"given":"Zhihua","family":"Jin"},{"given":"Shiyi","family":"Liu"},{"given":"Haotian","family":"Li"},{"given":"Xun","family":"Zhao"},{"given":"Huamin","family":"Qu"}],"first_public":"2024-07-03","date_note":"First arXiv version; the journal article appeared online in April 2025 and in the October 2025 issue.","date_source":"https://arxiv.org/abs/2407.03045","year":2025,"venue":"IEEE Transactions on Visualization and Computer Graphics","paper_url":"https://doi.org/10.1109/TVCG.2025.3557568","arxiv":"2407.03045","doi":"10.1109/TVCG.2025.3557568","category":"tool","relationship":"uses","dataset_version":"An earlier WildChat release containing more than 500,000 conversations, alongside LMSYS-Chat-1M.","summary":"Provides visual analysis at the group, conversation, and individual-turn levels to find jailbreak attempts in conversation logs. Analysts combine filtering, cluster views, conversation timelines, and comparisons with previously reported prompts; a case study uses WildChat to investigate multi-turn attempts after an initial model refusal.","finding":"The WildChat case study filters for multi-turn interactions and moderation scores to narrow the search to 503 conversations. Evaluation includes two case studies and interviews with nine researchers.","evidence_url":"https://arxiv.org/html/2407.03045v1","evidence_note":"Section III-C: datasets. Section VI-B: WildChat case study and filters. Section VI-C: expert evaluation.","bibtex_key":"jin2025jailbreakhunter","record_url":"https://yuntiandeng.com/wildchat/papers/jailbreakhunter/","relationship_label":"Uses WildChat","citation_title":"JailbreakHunter: A Visual Analytics Approach for Jailbreak Prompts Discovery From Large-Scale Human-LLM Conversational Datasets","citation_type":"article-journal","bibtex_type":"article","container_title":"IEEE Transactions on Visualization and Computer Graphics","citation_url":"https://doi.org/10.1109/TVCG.2025.3557568","citation_year":2025,"citation_note":"2025 journal article; the annotations use the July 2024 preprint.","pages":"7923–7937","volume":"31","issue":"10","publisher":"IEEE"}