{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,2,19]],"date-time":"2026-02-19T15:47:09Z","timestamp":1771516029406,"version":"3.50.1"},"reference-count":24,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,5,5]],"date-time":"2025-05-05T00:00:00Z","timestamp":1746403200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,5,5]],"date-time":"2025-05-05T00:00:00Z","timestamp":1746403200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,5,5]]},"DOI":"10.1109\/cscwd64889.2025.11033375","type":"proceedings-article","created":{"date-parts":[[2025,6,23]],"date-time":"2025-06-23T17:24:40Z","timestamp":1750699480000},"page":"273-278","source":"Crossref","is-referenced-by-count":2,"title":["DS-GCG: Enhancing LLM Jailbreaks with Token Suppression and Induction Dual-Strategy"],"prefix":"10.1109","author":[{"given":"Xuehai","family":"Tang","sequence":"first","affiliation":[{"name":"Institute of Information Engineering, Chinese Academy of Sciences"}]},{"given":"Xikang","family":"Yang","sequence":"additional","affiliation":[{"name":"Institute of Information Engineering, Chinese Academy of Sciences"}]},{"given":"Zhongjiang","family":"Yao","sequence":"additional","affiliation":[{"name":"Institute of Information Engineering, Chinese Academy of Sciences"}]},{"given":"Jie","family":"Wen","sequence":"additional","affiliation":[{"name":"Institute of Information Engineering, Chinese Academy of Sciences"}]},{"given":"Xi","family":"Zhou","sequence":"additional","affiliation":[{"name":"Institute of Information Engineering, Chinese Academy of Sciences"}]},{"given":"Jizhong","family":"Han","sequence":"additional","affiliation":[{"name":"Institute of Information Engineering, Chinese Academy of Sciences"}]},{"given":"Songlin","family":"Hu","sequence":"additional","affiliation":[{"name":"Institute of Information Engineering, Chinese Academy of Sciences"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Large language models for education: A survey and outlook","author":"Wang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref2","article-title":"Contrastive preference optimization: Pushing the boundaries of 11m performance in machine translation","author":"Haoran","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref3","article-title":"Exploring and evaluating hallucinations in 11m-powered code generation","author":"Fang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-47240-4_19"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1145\/3640543.3645143"},{"key":"ref6","first-page":"27730","article-title":"Training language models to follow instructions with human feedback","volume":"35","author":"Long","year":"2022","journal-title":"Advances in neural information processing systems"},{"key":"ref7","article-title":"Fine-tuning aligned language models compromises safety, even when users do not intend to!","author":"Xiangyu","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref8","article-title":"Autodan: Generating stealthy jail-break prompts on aligned large language models","author":"Xiaogeng","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref9","article-title":"Gptfuzzer: Red teaming large language models with auto-generated jailbreak prompts","author":"Jiahao","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref10","article-title":"Universal and transferable adversarial attacks on aligned language models","author":"Andy","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref11","article-title":"Autodan: Automatic and inter-pretable adversarial attacks on large language models","author":"Sicheng","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref12","article-title":"Jailbreaking black box large language models in twenty queries","author":"Patrick","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref13","article-title":"Tree of attacks: Jailbreaking black-box llms automatically","author":"Anay","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref14","article-title":"Mistral 7B","author":"Jiang","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref15","article-title":"Beavertails: Towards improved safety alignment of 11m via a human-preference dataset","volume":"36","author":"Jiaming","year":"2024","journal-title":"Ad-vances in Neural Information Processing Systems"},{"key":"ref16","article-title":"Safety Alignment Should Be Made More Than Just a Few Tokens Deep","author":"Xiangyu","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref17","article-title":"How johnny can persuade llms to jail-break them: Rethinking persuasion to challenge ai safety by humanizing llms","author":"Yi","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref18","article-title":"Foot In The Door: Understanding Large Language Model Jailbreaking via Cognitive Psychology","author":"Zhenhua","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref19","article-title":"Pal: Proxy-guided black-box attack on large language models","author":"Chawin","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref20","article-title":"Advprompter: Fast adaptive adversarial prompting for llms","author":"Anselm","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref21","article-title":"Llama guard: LIm-based input-output safeguard for human-ai conversations","author":"Hakan","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref22","article-title":"Scalable and transferable black-box jailbreaks for language models via persona modulation","author":"Rusheb","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref23","article-title":"Cognitive overload: J ailbreaking large language models with overloaded logical thinking","author":"Nan","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref24","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Hugo","year":"2023","journal-title":"arXiv preprint arXiv"}],"event":{"name":"2025 28th International Conference on Computer Supported Cooperative Work in Design (CSCWD)","location":"Compiegne, France","start":{"date-parts":[[2025,5,5]]},"end":{"date-parts":[[2025,5,7]]}},"container-title":["2025 28th International Conference on Computer Supported Cooperative Work in Design (CSCWD)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11033175\/11033221\/11033375.pdf?arnumber=11033375","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,6,24]],"date-time":"2025-06-24T06:05:19Z","timestamp":1750745119000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11033375\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,5]]},"references-count":24,"URL":"https:\/\/doi.org\/10.1109\/cscwd64889.2025.11033375","relation":{},"subject":[],"published":{"date-parts":[[2025,5,5]]}}}