{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,23]],"date-time":"2026-05-23T06:04:45Z","timestamp":1779516285025,"version":"3.53.1"},"reference-count":54,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,1,1]],"date-time":"2026-01-01T00:00:00Z","timestamp":1767225600000},"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":["62525201"],"award-info":[{"award-number":["62525201"]}],"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":["62132001"],"award-info":[{"award-number":["62132001"]}],"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":["62432001"],"award-info":[{"award-number":["62432001"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100005089","name":"Beijing Natural Science Foundation","doi-asserted-by":"publisher","award":["L247006"],"award-info":[{"award-number":["L247006"]}],"id":[{"id":"10.13039\/501100005089","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100005089","name":"Beijing Natural Science Foundation","doi-asserted-by":"publisher","award":["L257005"],"award-info":[{"award-number":["L257005"]}],"id":[{"id":"10.13039\/501100005089","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. on Image Process."],"published-print":{"date-parts":[[2026]]},"DOI":"10.1109\/tip.2026.3687073","type":"journal-article","created":{"date-parts":[[2026,5,5]],"date-time":"2026-05-05T20:05:03Z","timestamp":1778011503000},"page":"4907-4920","source":"Crossref","is-referenced-by-count":0,"title":["FinePruner: Unbiased Attention-Head-Level Fine-Grained Token Reduction for Efficient Inference of Large Vision-Language Models"],"prefix":"10.1109","volume":"35","author":[{"ORCID":"https:\/\/orcid.org\/0009-0007-0557-9647","authenticated-orcid":false,"given":"Zishuo","family":"Wang","sequence":"first","affiliation":[{"name":"Wangxuan Institute of Computer Technology, Peking University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-8526-3181","authenticated-orcid":false,"given":"Xiangtian","family":"Zheng","sequence":"additional","affiliation":[{"name":"Wangxuan Institute of Computer Technology, Peking University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-7658-3845","authenticated-orcid":false,"given":"Yuxin","family":"Peng","sequence":"additional","affiliation":[{"name":"Wangxuan Institute of Computer Technology, Peking University, Beijing, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"GPT-4o system card","author":"Hurst","year":"2024","journal-title":"arXiv:2410.21276"},{"key":"ref2","article-title":"Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context","author":"Team","year":"2024","journal-title":"arXiv:2403.05530"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref4","article-title":"Qwen2.5-VL technical report","volume-title":"arXiv:2502.13923","author":"Bai","year":"2025"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52733.2024.02283"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1723"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.00850"},{"key":"ref8","article-title":"Analyzing and boosting the power of fine-grained visual recognition for multi-modal large language models","author":"He","year":"2025","journal-title":"arXiv:2501.15140"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2021.3139234"},{"key":"ref10","article-title":"LLaVA-NeXT: Improved reasoning, OCR, and world knowledge","author":"Liu","year":"2024"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73004-7_2"},{"key":"ref12","article-title":"PyramidDrop: Accelerating your large vision-language models via pyramid visual redundancy reduction","author":"Xing","year":"2024","journal-title":"arXiv:2410.17247"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.00877"},{"key":"ref14","article-title":"Beyond text-visual attention: Exploiting visual cues for effective token pruning in VLMs","author":"Zhang","year":"2024","journal-title":"arXiv:2412.01818"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01843"},{"key":"ref16","article-title":"See what you are told: Visual attention sink in large multimodal models","author":"Kang","year":"2025","journal-title":"arXiv:2503.03321"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.802"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3126648"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2017.2774041"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2019.2921876"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2021.3135477"},{"key":"ref22","article-title":"The caltech-ucsd birds-200\u20132011 dataset","author":"Wah","year":"2011"},{"key":"ref23","article-title":"LLaMA: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv:2302.13971"},{"key":"ref24","volume-title":"Vicuna: An Open-Source Chatbot Impressing Gpt-4 With 90%* Chatgpt Quality","author":"Chiang et al","year":"2023"},{"key":"ref25","article-title":"Phi-3 technical report: A highly capable language model locally on your phone","volume-title":"arXiv:2404.14219","author":"Abdin","year":"2024"},{"key":"ref26","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Radford"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01100"},{"key":"ref28","article-title":"DINOv2: Learning robust visual features without supervision","author":"Oquab","year":"2023","journal-title":"arXiv:2304.07193"},{"key":"ref29","first-page":"19730","article-title":"BLIP-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Li"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-025-02491-7"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02777"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-naacl.242"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02325"},{"key":"ref34","article-title":"SparseVLM: Visual token sparsification for efficient vision-language model inference","author":"Zhang","year":"2024","journal-title":"arXiv:2410.04417"},{"key":"ref35","article-title":"[CLS] attention is all you need for training-free visual token pruning: Make VLM inference faster","author":"Zhang","year":"2024","journal-title":"arXiv:2412.01818v1"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01359"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.00878"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.1178"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.findings-emnlp.101"},{"key":"ref40","volume-title":"Interpreting GPT: The Logit Lens","year":"2020"},{"key":"ref41","article-title":"Towards interpreting visual information processing in vision-language models","author":"Neo","year":"2024","journal-title":"arXiv:2410.07149"},{"key":"ref42","article-title":"Longformer: The long-document transformer","author":"Beltagy","year":"2020","journal-title":"arXiv:2004.05150"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1145\/3773772"},{"key":"ref45","article-title":"MME: A comprehensive evaluation benchmark for multimodal large language models","author":"Fu","year":"2023","journal-title":"arXiv:2306.13394"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00686"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72658-3_13"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.52202\/068431-0182"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00851"},{"key":"ref50","article-title":"Fine-grained visual classification of aircraft","author":"Maji","year":"2013","journal-title":"arXiv:1306.5151"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW.2013.77"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/ICVGIP.2008.47"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2012.6248092"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.154"}],"container-title":["IEEE Transactions on Image Processing"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/83\/11355710\/11508157.pdf?arnumber=11508157","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,23]],"date-time":"2026-05-23T05:10:00Z","timestamp":1779513000000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11508157\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026]]},"references-count":54,"URL":"https:\/\/doi.org\/10.1109\/tip.2026.3687073","relation":{},"ISSN":["1057-7149","1941-0042"],"issn-type":[{"value":"1057-7149","type":"print"},{"value":"1941-0042","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026]]}}}