{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,10,3]],"date-time":"2026-10-03T20:37:56Z","timestamp":1791059876882,"version":"4.1.0"},"reference-count":94,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"10","license":[{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2026,10,1]],"date-time":"2026-10-01T00:00:00Z","timestamp":1790812800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"Ministry of Education - Singapore","award":["MOET2EP20221-0012"],"award-info":[{"award-number":["MOET2EP20221-0012"]}],"id":[{"id":"https:\/\/ror.org\/01kcva023","id-type":"ROR","asserted-by":"crossref"}]},{"name":"RIE2020 Industry Alignment Fund \u2013 Industry Collaboration Projects","award":["IAF-ICP"],"award-info":[{"award-number":["IAF-ICP"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Pattern Anal. Mach. Intell."],"published-print":{"date-parts":[[2026,10]]},"DOI":"10.1109\/tpami.2026.3697367","type":"journal-article","created":{"date-parts":[[2026,5,27]],"date-time":"2026-05-27T19:42:33Z","timestamp":1779910953000},"page":"12116-12131","source":"Crossref","is-referenced-by-count":7,"title":["Ego-R1: Agentic Chain-of-Tool-Thought for Ultra-Long Egocentric Video Reasoning"],"prefix":"10.1109","volume":"48","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9915-9377","authenticated-orcid":false,"given":"Shulin","family":"Tian","sequence":"first","affiliation":[{"name":"Nanyang Technological University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-3379-6103","authenticated-orcid":false,"given":"Ruiqi","family":"Wang","sequence":"additional","affiliation":[{"name":"Nanyang Technological University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongming","family":"Guo","sequence":"additional","affiliation":[{"name":"Shanghai AI Lab","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Penghao","family":"Wu","sequence":"additional","affiliation":[{"name":"Nanyang Technological University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0000-7936-2087","authenticated-orcid":false,"given":"Yuhao","family":"Dong","sequence":"additional","affiliation":[{"name":"Nanyang Technological University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Xiuying","family":"Wang","sequence":"additional","affiliation":[{"name":"Nanyang Technological University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Jingkang","family":"Yang","sequence":"additional","affiliation":[{"name":"Nanyang Technological University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1991-119X","authenticated-orcid":false,"given":"Hao","family":"Zhang","sequence":"additional","affiliation":[{"name":"Simon Fraser University","place":["Burnaby, BC, Canada"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Hongyuan","family":"Zhu","sequence":"additional","affiliation":[{"name":"A*STAR"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-4220-5958","authenticated-orcid":false,"given":"Ziwei","family":"Liu","sequence":"additional","affiliation":[{"name":"Nanyang Technological University"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","article-title":"GPT-4 technical report","author":"Achiam","year":"2023"},{"key":"ref2","article-title":"Claude 3.5 sonnet","year":"2024"},{"key":"ref3","article-title":"Qwen2. 5-vl technical report","author":"Bai","year":"2025"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.52202\/079017-1684"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v39i2.32214"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-025-02676-0"},{"key":"ref7","article-title":"LongVILA: Scaling long-context visual language models for long videos","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Chen","year":"2025"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01355"},{"key":"ref9","article-title":"SFT memorizes, RL generalizes: A comparative study of foundation model post-training","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Chu","year":"2025"},{"key":"ref10","first-page":"720","article-title":"Scaling egocentric vision: The epic-kitchens dataset","volume-title":"Proc. Eur. Conf. Comput. Vis.","author":"Damen"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1038\/s41586-025-09422-z"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.52202\/085713-4126"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01229"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.00847"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72670-5_5"},{"key":"ref16","article-title":"ReTool: Reinforcement learning for strategic tool use in LLMs","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Feng","year":"2026"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.52202\/085713-3310"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1177\/001316447303300309"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02245"},{"key":"ref20","article-title":"Gemini 3.1 pro preview","year":"2026"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72624-8_6"},{"key":"ref22","article-title":"Gemini 2.5: Our most intelligent AI model","year":"2025"},{"key":"ref23","article-title":"ToRA: A tool-integrated reasoning agent for mathematical problem solving","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Gou","year":"2024"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01842"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01834"},{"key":"ref26","article-title":"Vision-R1: Incentivizing reasoning capability in multimodal large language models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Huang","year":"2026"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW69036.2025.00614"},{"key":"ref28","article-title":"Search-R1: Training LLMs to reason and leverage search engines with reinforcement learning","volume-title":"Proc. Conf. Lang. Model.","author":"Jin","year":"2025"},{"key":"ref29","article-title":"VideoMem: Enhancing ultra-long video understanding via adaptive memory management","author":"Jin","year":"2025"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.00323"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-032-37627-5_35"},{"key":"ref32","article-title":"LLaVA-onevision: Easy visual task transfer","author":"Li","year":"2025","journal-title":"Trans. Mach. Learn. Res."},{"key":"ref33","first-page":"13876","article-title":"VideoChat: Chat-centric video understanding","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit.","author":"Li","year":"2024"},{"key":"ref34","article-title":"TORL: Scaling tool-integrated RL","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Li","year":"2026"},{"key":"ref35","article-title":"EgoToM: Benchmarking theory of mind reasoning from egocentric videos","author":"Li","year":"2025"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-acl.53"},{"key":"ref37","article-title":"Coarse correspondence elicit 3D spacetime understanding in multimodal language model","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Liu","year":"2024"},{"key":"ref38","article-title":"Oryx MLLM: On-demand spatial-temporal understanding at arbitrary resolution","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit.","author":"Liu","year":"2025"},{"key":"ref39","article-title":"Chain-of-spot: Interactive reasoning improves large vision-language models","author":"Liu","year":"2024"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.24894\/sprengglossarium_o01535"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.52202\/085713-5602"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.emnlp-main.564"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.acl-long.679"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2004"},{"key":"ref45","article-title":"Hello GPT-4o","year":"2024"},{"key":"ref46","article-title":"Introducing GPT-4.1 in the API","year":"2025"},{"key":"ref47","article-title":"Openai O3 and O4-mini system card","year":"2025"},{"key":"ref48","article-title":"TALM: Tool augmented language models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Parisi","year":"2025"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02226"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.52202\/085713-3524"},{"key":"ref51","article-title":"ToolLLM: Facilitating large language models to master 16000 real-world apis","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Qin","year":"2024"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.01942"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01762"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2997"},{"key":"ref55","article-title":"DeepSeekMath: Pushing the limits of mathematical reasoning in open language models","author":"Shao","year":"2024"},{"key":"ref56","article-title":"LongVU: Spatiotemporal adaptive compression for long video-language understanding","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Shen","year":"2025"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02436"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01725"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1688"},{"key":"ref60","article-title":"OpenThinkIMG: Learning to think with images via visual tool reinforcement learning","author":"Su","year":"2025"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.52202\/075280-3309"},{"key":"ref62","article-title":"Gemini 1.5: Unlocking multimodal understanding across millions of tokens of context","author":"Team","year":"2024"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.findings-acl.1247"},{"key":"ref64","article-title":"OTC: Optimal tool calls via reinforcement learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wang","year":"2025"},{"key":"ref65","article-title":"MathCoder: Seamless code integration in LLMs for enhanced mathematical reasoning","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Wang","year":"2024"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.52202\/085713-2243"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72989-8_4"},{"key":"ref68","article-title":"InternVideo2. 5: Empowering video MLLMs with long and rich context modeling","author":"Wang","year":"2025"},{"key":"ref69","article-title":"LifelongMemory: Leveraging LLMs for answering queries in long-form egocentric videos","author":"Wang","year":"2023"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.00311"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.52202\/068431-1800"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73414-4_26"},{"key":"ref73","article-title":"AutoGen: Enabling next-gen LLM applications via multi-agent conversation","volume-title":"Proc. Conf. Lang. Modeling","author":"Wu","year":"2024"},{"key":"ref74","article-title":"Grounded chain-of-thought for multimodal large language models","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit.","author":"Wu","year":"2026"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.00202"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01284"},{"key":"ref77","article-title":"Qwen2. 5-math technical report: Toward mathematical expert model via self-improvement","author":"Yang","year":"2024"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73232-4_2"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.02690"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.00229"},{"key":"ref81","article-title":"LongVT: Incentivizing \u2018thinking with long videos\u2019 via native tool calling","volume-title":"Proc. IEEE\/CVF Conf. Comput. Vis. Pattern Recognit.","author":"Yang","year":"2026"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.52202\/085713-5450"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.emnlp-main.158"},{"key":"ref84","article-title":"PRL: Process reward learning improves LLMs\u2019 reasoning ability and broadens the reasoning boundary","author":"Yao","year":"2026"},{"key":"ref85","article-title":"ReAct: Synergizing reasoning and acting in language models","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Yao"},{"key":"ref86","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.00802"},{"key":"ref87","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.00181"},{"key":"ref88","article-title":"Long context transfer from language to vision","author":"Zhang","year":"2024"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51701.2025.01918"},{"key":"ref90","article-title":"LLaVA-next: A strong zero-shot video understanding model","author":"Zhang","year":"2024"},{"key":"ref91","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2026.findings-acl.351"},{"key":"ref92","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2025.emnlp-main.22"},{"key":"ref93","article-title":"VideoAgent2: Enhancing the LLM-based agent system for long-form video understanding by uncertainty-aware cot","author":"Zhi","year":"2025"},{"key":"ref94","article-title":"Solving challenging math word problems using GPT-4 code interpreter with code-based self-verification","volume-title":"Proc. Int. Conf. Learn. Representations","author":"Zhou","year":"2024"}],"container-title":["IEEE Transactions on Pattern Analysis and Machine Intelligence"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/34\/11674301\/11535730.pdf?arnumber=11535730","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,9,24]],"date-time":"2026-09-24T17:11:20Z","timestamp":1790269880000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11535730\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,10]]},"references-count":94,"journal-issue":{"issue":"10"},"URL":"https:\/\/doi.org\/10.1109\/tpami.2026.3697367","relation":{},"ISSN":["0162-8828","2160-9292","1939-3539"],"issn-type":[{"value":"0162-8828","type":"print"},{"value":"2160-9292","type":"electronic"},{"value":"1939-3539","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,10]]}}}