{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T20:14:24Z","timestamp":1778530464374,"version":"3.51.4"},"reference-count":59,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"5","license":[{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T00:00:00Z","timestamp":1777593600000},"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":["62322212"],"award-info":[{"award-number":["62322212"]}],"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":["U25B2070"],"award-info":[{"award-number":["U25B2070"]}],"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":["U23A20387"],"award-info":[{"award-number":["U23A20387"]}],"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":["62536003"],"award-info":[{"award-number":["62536003"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"name":"Pengcheng Laboratory Research Project","award":["PCL2025AS14"],"award-info":[{"award-number":["PCL2025AS14"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Circuits Syst. Video Technol."],"published-print":{"date-parts":[[2026,5]]},"DOI":"10.1109\/tcsvt.2026.3652189","type":"journal-article","created":{"date-parts":[[2026,1,12]],"date-time":"2026-01-12T22:04:02Z","timestamp":1768255442000},"page":"5865-5876","source":"Crossref","is-referenced-by-count":0,"title":["Group-Relative Visual Discrimination Enhancement for Unlocking Intrinsic Capability of MLLMs"],"prefix":"10.1109","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-3948-7413","authenticated-orcid":false,"given":"Fang","family":"Peng","sequence":"first","affiliation":[{"name":"State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS), Institute of Automation, Chinese Academy of Sciences, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-5453-9755","authenticated-orcid":false,"given":"Xiaoshan","family":"Yang","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS), Institute of Automation, Chinese Academy of Sciences, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2197-9038","authenticated-orcid":false,"given":"Yaowei","family":"Wang","sequence":"additional","affiliation":[{"name":"Pengcheng Laboratory, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8343-9665","authenticated-orcid":false,"given":"Changsheng","family":"Xu","sequence":"additional","affiliation":[{"name":"State Key Laboratory of Multimodal Artificial Intelligence Systems (MAIS), Institute of Automation, Chinese Academy of Sciences, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"GPT-4 technical report","volume-title":"arXiv:2303.08774","author":"Achiam","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":"10.52202\/075280-2142"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref5","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":"ref6","doi-asserted-by":"publisher","DOI":"10.1145\/3711680"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.52202\/079017-3656"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3321501"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/3664647.3681071"},{"key":"ref10","article-title":"Exploring the reasoning abilities of multimodal large language models (MLLMs): A comprehensive survey on emerging trends in multimodal reasoning","author":"Wang","year":"2024","journal-title":"arXiv:2401.06805"},{"key":"ref11","article-title":"Insight-V: Exploring long-chain visual reasoning with multimodal large language models","author":"Dong","year":"2024","journal-title":"arXiv:2411.14432"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2020.3034233"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2023.3261659"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00914"},{"key":"ref15","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Radford","year":"2021"},{"key":"ref16","article-title":"RAR: Retrieving and ranking augmented MLLMs for visual recognition","author":"Liu","year":"2024","journal-title":"arXiv:2403.13805"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.01641"},{"key":"ref18","article-title":"MLLMs know where to look: Training-free perception of small visual details with multimodal LLMs","volume-title":"Proc. 13th Int. Conf. Learn. Represent.","author":"Zhang"},{"key":"ref19","first-page":"1","article-title":"Dinov2: Learning robust visual features without supervision","volume-title":"Trans. Mach. Learn. Res. J.","author":"Oquab","year":"Jan. 2024"},{"key":"ref20","article-title":"From training-free to adaptive: Empirical insights into MLLMs\u2019 understanding of detection information","author":"Jiao","year":"2024","journal-title":"arXiv:2401.17981"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2024.3521729"},{"key":"ref22","article-title":"Kosmos-2: Grounding multimodal large language models to the world","author":"Peng","year":"2023","journal-title":"arXiv:2306.14824"},{"key":"ref23","article-title":"Instruction tuning with GPT-4","author":"Peng","year":"2023","journal-title":"arXiv:2304.03277"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2024.3491176"},{"key":"ref25","article-title":"ViLBERT: Pretraining task-agnostic visiolinguistic representations for vision-and-language tasks","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Lu"},{"key":"ref26","article-title":"VisualBERT: A simple and performant baseline for vision and language","author":"Li","year":"2019","journal-title":"arXiv:1908.03557"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02484"},{"key":"ref28","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":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01227"},{"key":"ref30","article-title":"Qwen technical report","volume-title":"arXiv:2309.16609","author":"Bai","year":"2023"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/tcsvt.2025.3643469"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2025.3639574"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/tcsvt.2025.3619558"},{"key":"ref34","article-title":"Enhancing perception capabilities of multimodal LLMs with training-free fusion","author":"Chen","year":"2024","journal-title":"arXiv:2412.01289"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/tcsvt.2025.3643469"},{"key":"ref36","article-title":"Improving multi-modal large language model through boosting vision capabilities","author":"Sun","year":"2024","journal-title":"arXiv:2410.13733"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2024.3488181"},{"key":"ref38","article-title":"Token merging: Your ViT but faster","author":"Bolya","year":"2022","journal-title":"arXiv:2210.09461"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0969"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73004-7_2"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i2.32171"},{"key":"ref42","first-page":"4299","article-title":"Deep reinforcement learning from human preferences","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"Christiano"},{"key":"ref43","article-title":"Proximal policy optimization algorithms","author":"Schulman","year":"2017","journal-title":"arXiv:1707.06347"},{"key":"ref44","first-page":"27730","article-title":"Training language models to follow instructions with human feedback","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Ouyang"},{"key":"ref45","article-title":"Vision-r1: Incentivizing reasoning capability in multimodal large language models","author":"Huang","year":"2025","journal-title":"arXiv:2503.06749"},{"key":"ref46","article-title":"DeepSeek-r1: Incentivizing reasoning capability in LLMs via reinforcement learning","author":"Guo","year":"2025","journal-title":"arXiv:2501.12948"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01310"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.00245"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1007\/BF00992696"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2004.383"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2010.5539970"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2012.6248092"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10599-4_29"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72658-3_13"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.52202\/079017-0850"},{"key":"ref57","article-title":"GPT4 Vis: What can GPT-4 do for zero-shot visual recognition?","author":"Wu","year":"2023","journal-title":"arXiv:2311.15732"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-022-01653-1"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02249"}],"container-title":["IEEE Transactions on Circuits and Systems for Video Technology"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/76\/11511351\/11343853.pdf?arnumber=11343853","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,11]],"date-time":"2026-05-11T19:48:51Z","timestamp":1778528931000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11343853\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,5]]},"references-count":59,"journal-issue":{"issue":"5"},"URL":"https:\/\/doi.org\/10.1109\/tcsvt.2026.3652189","relation":{},"ISSN":["1051-8215","1558-2205"],"issn-type":[{"value":"1051-8215","type":"print"},{"value":"1558-2205","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,5]]}}}