{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T10:06:45Z","timestamp":1777889205793,"version":"3.51.4"},"reference-count":39,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"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":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.00290","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"3023-3033","source":"Crossref","is-referenced-by-count":0,"title":["SAUCE: Selective Concept Unlearning in Vision-Language Models with Sparse Autoencoders"],"prefix":"10.1109","author":[{"given":"Jiahui","family":"Geng","sequence":"first","affiliation":[{"name":"Mohamed bin Zayed University of Artificial Intelligence"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Qing","family":"Li","sequence":"additional","affiliation":[{"name":"Mohamed bin Zayed University of Artificial Intelligence"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Improving steering vectors by targeting sparse autoencoder features","author":"Chalnev","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref2","article-title":"Saeuron: Interpretable concept unlearning in diffusion models with sparse autoencoders","author":"Cywi\u0144ski","year":"2025","journal-title":"arXiv preprint arXiv"},{"key":"ref3","article-title":"How robust is google\u2019s bard to adversarial image attacks?","author":"Dong","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref4","article-title":"Unmemorization in large language models via self-distillation and deliberate imagination","author":"Dong","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.1058"},{"key":"ref6","article-title":"An image is worth 16x16 words: Transformers for image recognition at scale","volume-title":"International Conference on Learning Representations","author":"Dosovitskiy"},{"key":"ref7","article-title":"The llama 3 herd of models","author":"Dubey","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref8","article-title":"Who\u2019s harry potter? approximate unlearning in 11 ms","author":"Eldan","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref9","article-title":"Applying sparse autoencoders to unlearn knowledge in language models","volume":"2","author":"Farrell","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref10","article-title":"Mme: A comprehensive evaluation benchmark for multimodal large language models","author":"Fu","year":"2023","journal-title":"ArXiv, abs\/2306.13394"},{"key":"ref11","article-title":"Scaling and evaluating sparse autoencoders","volume-title":"The Thirteenth International Conference on Learning Representations","author":"Gao"},{"key":"ref12","article-title":"A comprehensive survey of machine unlearning techniques for large language models","volume":"1","author":"Geng","year":"2025","journal-title":"arXiv preprint arXiv"},{"key":"ref13","article-title":"Sparse autoencoders find highly interpretable features in language models","volume-title":"The Twelfth International Conference on Learning Representations","author":"Huben"},{"key":"ref14","article-title":"Sparse autoencoders find highly interpretable features in language models","volume-title":"The Twelfth International Conference on Learning Representations","author":"Huben"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.805"},{"key":"ref16","article-title":"Reversing the forget-retain objectives: An efficient LLM unlearning framework from logit difference","volume-title":"The Thirty-eighth Annual Conference on Neural Information Processing Systems","author":"Ji"},{"key":"ref17","volume-title":"Llava-next: Stronger llms supercharge multimodal capabilities in the wild","author":"Li","year":"2024"},{"key":"ref18","article-title":"Single image unlearning: Efficient machine unlearning in multimodal large language models","volume-title":"The Thirty-eighth Annual Conference on Neural Information Processing Systems","author":"Li"},{"issue":"3","key":"ref19","first-page":"243","article-title":"Continual learning and private unlearning","volume-title":"Conference on Lifelong Learning Agents","volume":"1","author":"Liu"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref22","first-page":"4105","article-title":"Protecting privacy in multimodal large language models with MLLMUbench","volume-title":"Proceedings of the 2025 Conference of the Nations of the Americas Chapter of the Association for Computational Linguistics: Human Language Technologies","volume":"1","author":"Liu","year":"2025"},{"key":"ref23","article-title":"Eight methods to evaluate robust unlearning in 11 ms","author":"Lynch","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref24","first-page":"5345","article-title":"Unveiling entity-level unlearning for large language models: A comprehensive analysis","volume-title":"Proceedings of the 31st International Conference on Computational Linguistics","author":"Ma"},{"key":"ref25","article-title":"Benchmarking vision language model unlearning via fictitious facial identity dataset","volume-title":"The Thirteenth International Conference on Learning Representations","volume":"1","author":"Ma"},{"key":"ref26","article-title":"Benchmarking vision language model unlearning via fictitious facial identity dataset","volume-title":"The Thirteenth International Conference on Learning Representations","author":"Ma"},{"key":"ref27","article-title":"TOFU: A task of fictitious unlearning for LLMs","volume-title":"First Conference on Language Modeling","volume":"3","author":"Maini"},{"key":"ref28","first-page":"2","volume-title":"Llama-3.2\u201311b-vision-instruct.","year":"2023"},{"key":"ref29","article-title":"A survey of machine unlearning","author":"Tam Nguyen","year":"2022","journal-title":"arXiv preprint arXiv"},{"key":"ref30","article-title":"Steering language model refusal with sparse autoencoders","volume":"2","author":"O\u2019Brien","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref31","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"International conference on machine learning","author":"Radford"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-015-0816-y"},{"key":"ref33","volume-title":"Vicuna: An open-source chatbot impressing gpt-4 with 90","year":"2023"},{"key":"ref34","first-page":"103619","article-title":"Private attribute inference from images with visionlanguage models","volume":"37","author":"T\u00f6mek\u00e7e","year":"2025","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref35","article-title":"Rkld: Reverse kl-divergence-based knowledge distillation for unlearning personal information in large language models","author":"Wang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.457"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.52202\/079017-3346"},{"key":"ref38","article-title":"Negative preference optimization: From catastrophic collapse to effective unlearning","volume-title":"First Conference on Language Modeling","author":"Zhang"},{"key":"ref39","article-title":"Catastrophic failure of LLM unlearning via quantization","volume-title":"The Thirteenth International Conference on Learning Representations","author":"Zhang"}],"event":{"name":"2025 IEEE\/CVF International Conference on Computer Vision (ICCV)","location":"Honolulu, HI, USA","start":{"date-parts":[[2025,10,19]]},"end":{"date-parts":[[2025,10,25]]}},"container-title":["2025 IEEE\/CVF International Conference on Computer Vision (ICCV)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11443115\/11443287\/11445313.pdf?arnumber=11445313","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T05:08:30Z","timestamp":1777612110000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11445313\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":39,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.00290","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}