{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,26]],"date-time":"2026-03-26T15:38:20Z","timestamp":1774539500453,"version":"3.50.1"},"reference-count":45,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"6","license":[{"start":{"date-parts":[[2025,6,1]],"date-time":"2025-06-01T00:00:00Z","timestamp":1748736000000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,6,1]],"date-time":"2025-06-01T00:00:00Z","timestamp":1748736000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,6,1]],"date-time":"2025-06-01T00:00:00Z","timestamp":1748736000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62306098"],"award-info":[{"award-number":["62306098"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["U23B2031"],"award-info":[{"award-number":["U23B2031"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Open Projects Program of State Key Laboratory of Multimodal Artificial Intelligence Systems"},{"DOI":"10.13039\/501100012226","name":"Fundamental Research Funds for the Central Universities","doi-asserted-by":"publisher","award":["JZ2024HGTB0256"],"award-info":[{"award-number":["JZ2024HGTB0256"]}],"id":[{"id":"10.13039\/501100012226","id-type":"DOI","asserted-by":"publisher"}]},{"name":"National Natural Science Foundation, Singapore under its AI Singapore Programme","award":["AISG2-RP-2021-022"],"award-info":[{"award-number":["AISG2-RP-2021-022"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Circuits Syst. Video Technol."],"published-print":{"date-parts":[[2025,6]]},"DOI":"10.1109\/tcsvt.2025.3526248","type":"journal-article","created":{"date-parts":[[2025,1,6]],"date-time":"2025-01-06T14:53:32Z","timestamp":1736175212000},"page":"5475-5488","source":"Crossref","is-referenced-by-count":3,"title":["Align Is Not Enough: Multimodal Universal Jailbreak Attack Against Multimodal Large Language Models"],"prefix":"10.1109","volume":"35","author":[{"ORCID":"https:\/\/orcid.org\/0009-0003-5621-6310","authenticated-orcid":false,"given":"Youze","family":"Wang","sequence":"first","affiliation":[{"name":"School of Computer Science and Information Engineering, Hefei University of Technology, Hefei, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0639-2012","authenticated-orcid":false,"given":"Wenbo","family":"Hu","sequence":"additional","affiliation":[{"name":"School of Computer Science and Information Engineering, Hefei University of Technology, Hefei, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1299-683X","authenticated-orcid":false,"given":"Yinpeng","family":"Dong","sequence":"additional","affiliation":[{"name":"Department of Computer Science and Technology, Tsinghua University, Beijing, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0903-9131","authenticated-orcid":false,"given":"Jing","family":"Liu","sequence":"additional","affiliation":[{"name":"Institute of Automation, Chinese Academy of Science, Beijing, China"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7374-8739","authenticated-orcid":false,"given":"Hanwang","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Engineering, Nanyang Technological University, Jurong West, Singapore"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5461-3986","authenticated-orcid":false,"given":"Richang","family":"Hong","sequence":"additional","affiliation":[{"name":"School of Computer Science and Information Engineering, Hefei University of Technology, Hefei, China"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Gpt-4 technical report","volume-title":"arXiv:2303.08774","author":"OpenAI","year":"2023"},{"key":"ref2","article-title":"Gemini: A family of highly capable multimodal models","author":"Team","year":"2023","journal-title":"arXiv:2312.11805"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1093\/nsr\/nwae403"},{"key":"ref4","first-page":"1","article-title":"Visual instruction tuning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Liu"},{"key":"ref5","article-title":"VideoChat: Chat-centric video understanding","author":"Li","year":"2023","journal-title":"arXiv:2305.06355"},{"key":"ref6","first-page":"18090","article-title":"Pengi: An audio language model for audio tasks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Deshmukh"},{"key":"ref7","first-page":"1","article-title":"LLaVA-med: Training a large language-and-vision assistant for biomedicine in one day","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Li"},{"key":"ref8","article-title":"Universal and transferable adversarial attacks on aligned language models","author":"Zou","year":"2023","journal-title":"arXiv:2307.15043"},{"key":"ref9","article-title":"GPT-4 is too smart to be safe: Stealthy chat with LLMs via cipher","author":"Yuan","year":"2023","journal-title":"arXiv:2308.06463"},{"key":"ref10","article-title":"Jailbreaking black box large language models in twenty queries","author":"Chao","year":"2023","journal-title":"arXiv:2310.08419"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i19.30150"},{"key":"ref12","article-title":"Visual instruction tuning","author":"Liu","year":"2023","journal-title":"arXiv:2304.08485"},{"key":"ref13","article-title":"Yi: Open foundation models by 01.AI","author":"Young","year":"2024","journal-title":"arXiv:2403.04652"},{"key":"ref14","article-title":"MiniGPT-v2: Large language model as a unified interface for vision-language multi-task learning","author":"Chen","year":"2023","journal-title":"arXiv:2310.09478"},{"key":"ref15","article-title":"MiniGPT-4: Enhancing vision-language understanding with advanced large language models","author":"Zhu","year":"2023","journal-title":"arXiv:2304.10592"},{"key":"ref16","first-page":"1","article-title":"InstructBLIP: Towards general-purpose vision-language models with instruction tuning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Dai"},{"key":"ref17","article-title":"Qwen2-VL: Enhancing vision-language model\u2019s perception of the world at any resolution","author":"Wang","year":"2024","journal-title":"arXiv:2409.12191"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01239"},{"key":"ref19","article-title":"MiniCPM-V: A GPT-4 V level MLLM on your phone","author":"Yao","year":"2024","journal-title":"arXiv:2408.01800"},{"key":"ref20","article-title":"CogVLM: Visual expert for pretrained language models","author":"Wang","year":"2023","journal-title":"arXiv:2311.03079"},{"key":"ref21","article-title":"COLD-attack: Jailbreaking LLMs with stealthiness and controllability","author":"Guo","year":"2024","journal-title":"arXiv:2402.08679"},{"key":"ref22","article-title":"Jailbreaking attack against multimodal large language model","author":"Niu","year":"2024","journal-title":"arXiv:2402.02309"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3307150"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3263054"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2024.3351680"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3345167"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3276442"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1145\/3503161.3547801"},{"key":"ref29","article-title":"Exploring transferability of multimodal adversarial samples for vision-language pre-training models with contrastive learning","author":"Wang","year":"2023","journal-title":"arXiv:2308.12636"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/TIFS.2024.3460387"},{"key":"ref31","first-page":"1","article-title":"An unforgeable publicly verifiable watermark for large language models","volume-title":"Proc. 12th Int. Conf. Learn. Represent.","author":"Liu"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1016\/j.cose.2023.103102"},{"key":"ref33","article-title":"MultiTrust: A comprehensive benchmark towards trustworthy multimodal large language models","author":"Zhang","year":"2024","journal-title":"arXiv:2406.07057"},{"key":"ref34","article-title":"Visually adversarial attacks and defenses in the physical world: A survey","author":"Wei","year":"2022","journal-title":"arXiv:2211.01671"},{"key":"ref35","article-title":"How to bridge the gap between modalities: A comprehensive survey on multimodal large language model","author":"Song","year":"2023","journal-title":"arXiv:2311.07594"},{"key":"ref36","article-title":"Towards deep learning models resistant to adversarial attacks","author":"Madry","year":"2017","journal-title":"arXiv:1706.06083"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i19.30150"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00196"},{"key":"ref39","first-page":"1","article-title":"PyTorch: An imperative style, high-performance deep learning library","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Paszke"},{"key":"ref40","first-page":"1","article-title":"Jailbreak in pieces: Compositional adversarial attacks on multi-modal language models","volume-title":"Proc. 12th Int. Conf. Learn. Represent.","author":"Shayegani"},{"key":"ref41","article-title":"Jailbreak vision language models via bi-modal adversarial prompt","author":"Ying","year":"2024","journal-title":"arXiv:2406.04031"},{"key":"ref42","article-title":"SmoothLLM: Defending large language models against jailbreaking attacks","author":"Robey","year":"2023","journal-title":"arXiv:2310.03684"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.naacl-long.92"},{"key":"ref44","article-title":"Refuse whenever you feel unsafe: Improving safety in LLMs via decoupled refusal training","author":"Yuan","year":"2024","journal-title":"arXiv:2407.09121"},{"key":"ref45","article-title":"Efficient adversarial training in LLMs with continuous attacks","author":"Xhonneux","year":"2024","journal-title":"arXiv:2405.15589"}],"container-title":["IEEE Transactions on Circuits and Systems for Video Technology"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/76\/11027896\/10829683.pdf?arnumber=10829683","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,1,5]],"date-time":"2026-01-05T18:42:04Z","timestamp":1767638524000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10829683\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6]]},"references-count":45,"journal-issue":{"issue":"6"},"URL":"https:\/\/doi.org\/10.1109\/tcsvt.2025.3526248","relation":{},"ISSN":["1051-8215","1558-2205"],"issn-type":[{"value":"1051-8215","type":"print"},{"value":"1558-2205","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,6]]}}}