{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,11]],"date-time":"2025-12-11T19:08:26Z","timestamp":1765480106139,"version":"3.48.0"},"reference-count":47,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"24","license":[{"start":{"date-parts":[[2025,12,15]],"date-time":"2025-12-15T00:00:00Z","timestamp":1765756800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,12,15]],"date-time":"2025-12-15T00:00:00Z","timestamp":1765756800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,12,15]],"date-time":"2025-12-15T00:00:00Z","timestamp":1765756800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Key Research and Development Program of China","award":["2024YFB3311802"],"award-info":[{"award-number":["2024YFB3311802"]}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62172377"],"award-info":[{"award-number":["62172377"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Taishan Scholars Program of Shandong Province","award":["tsqn202312102"],"award-info":[{"award-number":["tsqn202312102"]}]},{"name":"Startup Research Foundation for Distinguished Scholars","award":["202112016"],"award-info":[{"award-number":["202112016"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Internet Things J."],"published-print":{"date-parts":[[2025,12,15]]},"DOI":"10.1109\/jiot.2025.3599199","type":"journal-article","created":{"date-parts":[[2025,8,18]],"date-time":"2025-08-18T19:46:41Z","timestamp":1755546401000},"page":"51892-51901","source":"Crossref","is-referenced-by-count":0,"title":["Fast and Controllable Bias-Guided Jailbreak Attack on Large Language Models"],"prefix":"10.1109","volume":"12","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-1863-2802","authenticated-orcid":false,"given":"Zi","family":"Kang","sequence":"first","affiliation":[{"name":"College of Computer Science and Technology, Ocean University of China, Qingdao, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7326-5796","authenticated-orcid":false,"given":"Hui","family":"Xia","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Ocean University of China, Qingdao, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4117-2656","authenticated-orcid":false,"given":"Rui","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Cybersecurity, Qufu Normal University, Jining, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0006-3911-8331","authenticated-orcid":false,"given":"Xiaoxue","family":"Song","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Ocean University of China, Qingdao, China"}]},{"ORCID":"https:\/\/orcid.org\/0009-0007-0742-8498","authenticated-orcid":false,"given":"Le","family":"Li","sequence":"additional","affiliation":[{"name":"College of Computer Science and Technology, Ocean University of China, Qingdao, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5825-2241","authenticated-orcid":false,"given":"Chunqiang","family":"Hu","sequence":"additional","affiliation":[{"name":"School of Big Data and Software Engineering, Chongqing University, Chongqing, China"}]}],"member":"263","reference":[{"key":"ref1","first-page":"1","article-title":"Emergent abilities of large language models","author":"Wei","year":"2022","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref2","first-page":"1","article-title":"Jailbroken: How does LLM safety training fail?","volume-title":"Proc. 37th Conf. Neural Inf. Process. Syst.","author":"Wei"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.773"},{"key":"ref4","article-title":"From noise to clarity: Unraveling the adversarial suffix of large language model attacks via translation of text embeddings","author":"Wang","year":"2024","journal-title":"arXiv:2402.16006"},{"key":"ref5","article-title":"Jailbreaking leading safety-aligned LLMs with simple adaptive attacks","author":"Andriushchenko","year":"2024","journal-title":"arXiv:2404.02151"},{"key":"ref6","article-title":"Tree of attacks: Jailbreaking black-box LLMs automatically","author":"Mehrotra","year":"2023","journal-title":"arXiv:2312.02119"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1145\/3543507.3583306"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/TC.2023.3280133"},{"key":"ref9","article-title":"Ignore previous prompt: Attack techniques for language models","author":"Perez","year":"2022","journal-title":"arXiv:2211.09527"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1145\/3605764.3623985"},{"key":"ref11","article-title":"Universal and transferable adversarial attacks on aligned language models","author":"Zou","year":"2023","journal-title":"arXiv:2307.15043"},{"key":"ref12","first-page":"1","article-title":"AutoDAN: Generating stealthy jailbreak prompts on aligned large language models","volume-title":"Proc. 12th Int. Conf. Learn. Represent.","author":"Liu"},{"key":"ref13","first-page":"1","article-title":"COLD-attack: Jailbreaking LLMs with stealthiness and controllability","volume-title":"Proc. 41st Int. Conf. Mach. Learn.","author":"Guo"},{"key":"ref14","first-page":"1","article-title":"Hard prompts made easy: Gradient-based discrete optimization for prompt tuning and discovery","volume-title":"Proc. 37th Conf. Neural Inf. Process. Syst.","author":"Wen"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-short.18"},{"key":"ref16","article-title":"Yuan 1.0: Large-scale pre-trained language model in zero-shot and few-shot learning","author":"Wu","year":"2021","journal-title":"arXiv:2110.04725"},{"issue":"8","key":"ref17","first-page":"9","article-title":"Language models are unsupervised multitask learners","volume":"1","author":"Radford","year":"2019","journal-title":"OpenAI Blog"},{"key":"ref18","first-page":"186","article-title":"Language models are few-shot learners","volume-title":"Proc. 34th Conf Neural Inf. Process. Syst.","author":"Mann"},{"key":"ref19","article-title":"GPT-4 technical report","volume-title":"arXiv:2303.08774","author":"Achiam","year":"2023"},{"key":"ref20","article-title":"Evaluating large language models trained on code","author":"Chen","year":"2021","journal-title":"arXiv:2107.03374"},{"key":"ref21","article-title":"WebGPT: Browser-assisted question-answering with human feedback","author":"Nakano","year":"2021","journal-title":"arXiv:2112.09332"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.1810.04805"},{"key":"ref23","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv:2302.13971"},{"issue":"6","key":"ref24","first-page":"7","article-title":"Alpaca: A strong, replicable instruction-following model","volume":"3","author":"Taori","year":"2023","journal-title":"Stanford Center Res. Found. Models"},{"key":"ref25","first-page":"6","article-title":"Koala: A dialogue model for academic research","volume":"1","author":"Geng","year":"2023","journal-title":"Blog Post"},{"key":"ref26","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"GenAI","year":"2023","journal-title":"arXiv:2307.09288"},{"key":"ref27","article-title":"Mistral 7b","author":"Jiang","year":"2023","journal-title":"arXiv:2310.06825"},{"key":"ref28","article-title":"Code llama: Open foundation models for code","author":"Grattafiori","year":"2023","journal-title":"arXiv:2308.12950"},{"key":"ref29","first-page":"42661","article-title":"Focused transformer: Contrastive training for context scaling","volume-title":"Proc. 37th Conf. Neural Inf. Process. Syst.","volume":"36","author":"Tworkowski"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.352"},{"key":"ref31","article-title":"Pathseeker: Exploring LLM security vulnerabilities with a reinforcement learning-based jailbreak approach","author":"Lin","year":"2024","journal-title":"arXiv:2409.14177"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-emnlp.813"},{"key":"ref33","first-page":"1","article-title":"Are aligned neural networks adversarially aligned?","volume-title":"Proc. 37th Conf. Neural Inf. Process. Syst.","author":"Carlini"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/3658644.3670388"},{"key":"ref35","article-title":"A practical trigger-free backdoor attack on neural networks","author":"Wang","year":"2024","journal-title":"arXiv:2408.11444"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2024.3486218"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/TMC.2024.3521216"},{"key":"ref38","article-title":"AutoDAN: Automatic and interpretable adversarial attacks on large language models","author":"Zhu","year":"2023","journal-title":"arXiv:2310.15140"},{"key":"ref39","first-page":"1","article-title":"Catastrophic jailbreak of open-source LLMs via exploiting generation","volume-title":"Proc. 12th Int. Conf. Learn. Represent.","author":"Huang"},{"key":"ref40","first-page":"1","article-title":"Fine-tuning aligned language models compromises safety, even when users do not intend to!","volume-title":"Proc. 12th Int. Conf. Learn. Represent.","author":"Qi"},{"key":"ref41","article-title":"Prompt injection attack against LLM-integrated applications","author":"Liu","year":"2023","journal-title":"arXiv:2306.05499"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.3233\/AIC-230279"},{"volume-title":"Vicuna: An Open-Source Chatbot Impressing GPT-4 with 90%* ChatGPT Quality","year":"2023","author":"Chiang","key":"ref43"},{"key":"ref44","first-page":"1","article-title":"QLORA: Efficient finetuning of quantized LLMs","volume-title":"Proc. 37th Conf. Neural Inf. Process. Syst.","volume":"36","author":"Dettmers"},{"key":"ref45","article-title":"Baseline defenses for adversarial attacks against aligned language models","author":"Jain","year":"2023","journal-title":"arXiv:2309.00614"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.3115\/1073083.1073135"},{"key":"ref47","first-page":"1","article-title":"BERTScore: Evaluating text generation with BERT","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Zhang"}],"container-title":["IEEE Internet of Things Journal"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6488907\/11293846\/11126090.pdf?arnumber=11126090","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,12,11]],"date-time":"2025-12-11T18:48:20Z","timestamp":1765478900000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11126090\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12,15]]},"references-count":47,"journal-issue":{"issue":"24"},"URL":"https:\/\/doi.org\/10.1109\/jiot.2025.3599199","relation":{},"ISSN":["2327-4662","2372-2541"],"issn-type":[{"type":"electronic","value":"2327-4662"},{"type":"electronic","value":"2372-2541"}],"subject":[],"published":{"date-parts":[[2025,12,15]]}}}