{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,6,19]],"date-time":"2026-06-19T15:49:32Z","timestamp":1781884172215,"version":"3.54.5"},"reference-count":64,"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\/iccv51701.2025.01939","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"20857-20867","source":"Crossref","is-referenced-by-count":3,"title":["Beyond Text-Visual Attention: Exploiting Visual Cues for Effective Token Pruning in VLMs"],"prefix":"10.1109","author":[{"given":"Qizhe","family":"Zhang","sequence":"first","affiliation":[{"name":"School of Computer Science, Peking University,State Key Laboratory for Multimedia Information Processing"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Aosong","family":"Cheng","sequence":"additional","affiliation":[{"name":"School of Computer Science, Peking University,State Key Laboratory for Multimedia Information Processing"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ming","family":"Lu","sequence":"additional","affiliation":[{"name":"School of Computer Science, Peking University,State Key Laboratory for Multimedia Information Processing"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Renrui","family":"Zhang","sequence":"additional","affiliation":[{"name":"CUHK MMLab"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhiyong","family":"Zhuo","sequence":"additional","affiliation":[{"name":"School of Computer Science, Peking University,State Key Laboratory for Multimedia Information Processing"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jiajun","family":"Cao","sequence":"additional","affiliation":[{"name":"School of Computer Science, Peking University,State Key Laboratory for Multimedia Information Processing"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shaobo","family":"Guo","sequence":"additional","affiliation":[{"name":"ByteDance"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qi","family":"She","sequence":"additional","affiliation":[{"name":"ByteDance"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Shanghang","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Computer Science, Peking University,State Key Laboratory for Multimedia Information Processing"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Gpt-4 technical report","author":"Achiam","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref2","article-title":"Mc-llava: Multi-concept personalized vision-language model","author":"An","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref3","article-title":"Unictokens: Boosting personalized understanding and generation via unified concept tokens","author":"An","year":"2025","journal-title":"arXiv preprint arXiv"},{"key":"ref4","article-title":"Qwen technical report","author":"Bai","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref5","article-title":"Qwen-vl: A frontier large vision-language model with versatile abilities","author":"Bai","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref6","article-title":"Token merging: Your vit but faster","author":"Bolya","year":"2023","journal-title":"ICLR"},{"key":"ref7","article-title":"Matryoshka multimodal models","author":"Cai","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52734.2025.01848"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00951"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.52202\/079017-2353"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73004-7_2"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.02283"},{"key":"ref13","first-page":"4271","article-title":"Funnel-transformer: Filtering out sequential redundancy for efficient language processing","volume":"33","author":"Dai","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1189"},{"key":"ref15","article-title":"Vision transformers need registers","author":"Darcet","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref16","article-title":"An image is worth $16 \\times 16$ words: Transformers for image recognition at scale","author":"Dosovitskiy","journal-title":"arXiv preprint arXiv"},{"key":"ref17","author":"Fu","year":"2024","journal-title":"Mme: A comprehensive evaluation benchmark for multimodal large language models"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.670"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73010-8_23"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00380"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-emnlp.338"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.52202\/079017-1588"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01274"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-long.602"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00686"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.149"},{"key":"ref27","article-title":"Mistral 7b","author":"Albert","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref28","first-page":"1973019742","article-title":"Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","volume-title":"In International conference on machine learning","author":"Li"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-025-02491-7"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.emnlp-main.20"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72952-2_19"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.342"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref34","author":"Liu","year":"2024","journal-title":"Llava-next: Improved reasoning, ocr, and world knowledge"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.52202\/079017"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72658-3_13"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.52202\/068431-0182"},{"key":"ref38","article-title":"Feast your eyes: Mixture-ofresolution adaptation for multimodal large language models","author":"Luo","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.679"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.acl-long.353"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.52202\/068431-2011"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1145\/3725843.3756072"},{"key":"ref43","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"In International conference on machine learning","author":"Radford"},{"key":"ref44","article-title":"Compressive transformers for long-range sequence modelling","author":"Jack","year":"2019","journal-title":"arXiv preprint arXiv"},{"key":"ref45","article-title":"Llava-prumerge: Adaptive token reduction for efficient large multimodal models","author":"Shang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00851"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2023.127063"},{"key":"ref48","article-title":"Gemini: a family of highly capable multimodal models","author":"Team","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.00914"},{"key":"ref50","article-title":"Llama: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref51","article-title":"Llama 2: Open foundation and fine-tuned chat models","author":"Touvron","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref52","article-title":"Attention is all you need","author":"Vaswani","year":"2017","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref53","article-title":"[cls] token tells everything needed for training-free efficient mllms","author":"Wang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.52202\/079017-3860"},{"key":"ref55","article-title":"Pyramiddrop: Accelerating your large vision-language models via pyramid visual redundancy reduction","author":"Xing","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1145\/3123266.3123427"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01843"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i21.34366"},{"key":"ref59","article-title":"Mm -vet: Evaluating large multimodal models for integrated capabilities","author":"Yu","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33019127"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01100"},{"key":"ref62","article-title":"Llava-mini: Efficient image and video large multimodal models with one vision token","author":"Zhang","year":"2025","journal-title":"arXiv preprint arXiv"},{"key":"ref63","article-title":"Sparsevlm: Visual token sparsification for efficient vision-language model inference","author":"Zhang","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01845"}],"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\/11444044.pdf?arnumber=11444044","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T05:28:46Z","timestamp":1777613326000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11444044\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":64,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.01939","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}