{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,20]],"date-time":"2026-05-20T20:54:12Z","timestamp":1779310452059,"version":"3.51.4"},"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"}],"funder":[{"DOI":"10.13039\/501100012166","name":"National Key R&D Program of China","doi-asserted-by":"publisher","award":["2022ZD0161600"],"award-info":[{"award-number":["2022ZD0161600"]}],"id":[{"id":"10.13039\/501100012166","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100023542","name":"Centre for Perceptual and Interactive Intelligence (CPII) Ltd","doi-asserted-by":"publisher","id":[{"id":"10.13039\/501100023542","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.00197","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"2034-2044","source":"Crossref","is-referenced-by-count":7,"title":["Visual-RFT: Visual Reinforcement Fine-Tuning"],"prefix":"10.1109","author":[{"given":"Ziyu","family":"Liu","sequence":"first","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zeyi","family":"Sun","sequence":"additional","affiliation":[{"name":"Shanghai Jiao Tong University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuhang","family":"Zang","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaoyi","family":"Dong","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yuhang","family":"Cao","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Haodong","family":"Duan","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Dahua","family":"Lin","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jiaqi","family":"Wang","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Lmrl gym: Benchmarks for multi-turn reinforcement learning with language models","author":"Abdulhai","year":"2023","journal-title":"arXiv preprint"},{"key":"ref2","article-title":"Internlm2 technical report","author":"Cai","year":"2024","journal-title":"arXiv preprint"},{"key":"ref3","article-title":"Grounding large language models in interactive environments with online reinforcement learning","author":"Carta","year":"2023","journal-title":"ICLR"},{"key":"ref4","article-title":"Deepseek-r1: Incentivizing reasoning capability in 11 ms via reinforcement learning","author":"Guo","year":"2025","journal-title":"arXiv preprint"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00550"},{"key":"ref6","article-title":"Qwen2","author":"Hui","year":"2024","journal-title":"5-coder technical report"},{"key":"ref7","author":"Jaech","year":"2024","journal-title":"Openai o1 system card"},{"key":"ref8","article-title":"Preference optimization for reasoning with pseudo feedback","author":"Jiao","year":"2024","journal-title":"arXiv preprint"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01832"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01394"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW.2013.77"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00915"},{"key":"ref14","article-title":"T \\\u201c ulu 3: Pushing frontiers in open language model posttraining","author":"Lambert","year":"2024","journal-title":"arXiv preprint"},{"key":"ref15","article-title":"Llava-onevision: Easy visual task transfer","author":"Li","year":"2024","journal-title":"arXiv preprint"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00682"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10602-1_48"},{"key":"ref18","article-title":"Deepseek-v3 technical report","author":"Liu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref19","article-title":"Skywork-Reward: Bag of tricks for reward modeling in 11 ms","author":"Yuhao Liu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72970-6_3"},{"key":"ref21","article-title":"Mia-dpo: Multi-image augmented direct preference optimization for large vision-language models","author":"Liu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.410"},{"key":"ref23","article-title":"Fine-grained visual classi-fication of aircraft","author":"Maji","year":"2013","journal-title":"arXiv preprint"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/ICVGIP.2008.47"},{"key":"ref25","article-title":"OpenAI","year":"2024","journal-title":"Hello gpt-4o"},{"key":"ref26","article-title":"OpenAI","year":"2025","journal-title":"Openai o3-mini system card"},{"key":"ref27","article-title":"Training language models to follow instructions with human feedback","author":"Ouyang","year":"2022","journal-title":"In NeurIPS"},{"key":"ref28","article-title":"Training language models to follow instructions with human feedback","author":"Ouyang","year":"2022","journal-title":"NeurIPS"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2012.6248092"},{"key":"ref30","article-title":"Is reinforcement learning (not) for natural language processing: Benchmarks, baselines, and building blocks for natural language policy optimization","author":"Ramamurthy","year":"2023","journal-title":"ICLR"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01236"},{"key":"ref32","author":"Schulman","year":"2017","journal-title":"Proximal policy optimization algorithms"},{"key":"ref33","article-title":"Deepseekmath: Pushing the limits of mathematical reasoning in open language models","author":"Shao","year":"2024","journal-title":"arXiv preprint"},{"key":"ref34","article-title":"Offline RL for natural language generation with implicit language q learning","author":"Victor Snell","year":"2023","journal-title":"ICLR"},{"key":"ref35","article-title":"Christiano","volume":"3","author":"Stiennon","year":"2022","journal-title":"Learning to summarize with human feedback. In NeurIPS"},{"key":"ref36","article-title":"Aligning large multimodal models with factually augmented rlhf","author":"Sun","year":"2023","journal-title":"arXiv preprint"},{"key":"ref37","article-title":"Aligning large multimodal models with factually augmented rlhf","author":"Sun","year":"2023","journal-title":"arXiv preprint"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-acl.775"},{"key":"ref39","article-title":"Kimi k1. 5: Scaling reinforcement learning with 11 ms","author":"Team","year":"2025","journal-title":"arXiv preprint"},{"key":"ref40","article-title":"Qwen2-vl: Enhancing vision-language model\u2019s perception of the world at any resolution","author":"Wang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1145\/3664647.3680642"},{"key":"ref42","article-title":"Clipself: Vision transformer distills itself for open-vocabulary dense prediction","volume":"abs\/2310.01403","author":"Wu","year":"2023","journal-title":"ArXiv"},{"key":"ref43","article-title":"Qwen2. 5-math technical report: Toward mathematical expert model via self-improvement","author":"Yang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref44","article-title":"React: Synergizing reasoning and acting in language models","author":"Yao","year":"2023","journal-title":"ICLR"},{"key":"ref45","article-title":"Internlm-math: Open math large language models toward verifiable reasoning","author":"Ying","year":"2024","journal-title":"arXiv preprint"},{"key":"ref46","article-title":"RIHF-V: Towards trustworthy mllms via behavior alignment from fine-grained correctional human feedback","author":"Yu","year":"2024","journal-title":"CVPR"},{"key":"ref47","article-title":"RLAIF-V: Aligning mllms through open-source ai feedback for super gpt-4v trustworthiness","author":"Yu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-024-02214-4"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.340"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.01416"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.acl-long.771"},{"key":"ref52","article-title":"Internlm-xcomposer-2.5: A versatile large vision language model supporting long-contextual input and output","author":"Zhang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref53","article-title":"Detect every thing with few examples","volume":"abs\/2309.12969","author":"Zhang","year":"2023","journal-title":"ArXiv"},{"key":"ref54","article-title":"o1-coder: an o1 replication for coding","author":"Zhang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref55","article-title":"Beyond hallucinations: Enhancing lvlms through hallucination-aware direct preference optimization","author":"Zhao","year":"2023","journal-title":"arXiv preprint"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-022-01653-1"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01631"},{"key":"ref58","author":"Zhou","year":"2025","journal-title":"Inpk: Infusing prior knowledge into prompt for vision-language models"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20077-9_21"},{"key":"ref60","article-title":"Aligning modalities in vision large language models via preference fine-tuning","author":"Zhou","year":"2024","journal-title":"arXiv preprint"},{"key":"ref61","article-title":"Aligning modalities in vision large language models via preference fine-tuning","author":"Zhou","year":"2024","journal-title":"arXiv preprint"},{"key":"ref62","article-title":"Archer: Training language model agents via hierarchical multi-turn rl","author":"Zhou","year":"2024","journal-title":"ICML"},{"key":"ref63","author":"Ziegler","year":"2019","journal-title":"Fine-tuning language models from human preferences"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01451"}],"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\/11444749.pdf?arnumber=11444749","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T05:07:13Z","timestamp":1777612033000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11444749\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":64,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.00197","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}