{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,11,7]],"date-time":"2026-11-07T07:40:27Z","timestamp":1794037227454,"version":"4.8.0"},"reference-count":83,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,6,10]],"date-time":"2025-06-10T00:00:00Z","timestamp":1749513600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,6,10]],"date-time":"2025-06-10T00:00:00Z","timestamp":1749513600000},"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":[[2025,6,10]]},"DOI":"10.1109\/cvpr52734.2025.00166","type":"proceedings-article","created":{"date-parts":[[2025,8,13]],"date-time":"2025-08-13T17:26:42Z","timestamp":1755106002000},"page":"1702-1713","source":"Crossref","is-referenced-by-count":112,"title":["CoT-VLA: Visual Chain-of-Thought Reasoning for Vision-Language-Action Models"],"prefix":"10.1109","author":[{"given":"Qingqing","family":"Zhao","sequence":"first","affiliation":[{"name":"NVIDIA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yao","family":"Lu","sequence":"additional","affiliation":[{"name":"NVIDIA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Moo Jin","family":"Kim","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zipeng","family":"Fu","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhuoyang","family":"Zhang","sequence":"additional","affiliation":[{"name":"MIT"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yecheng","family":"Wu","sequence":"additional","affiliation":[{"name":"NVIDIA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Zhaoshuo","family":"Li","sequence":"additional","affiliation":[{"name":"NVIDIA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Qianli","family":"Ma","sequence":"additional","affiliation":[{"name":"NVIDIA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Song","family":"Han","sequence":"additional","affiliation":[{"name":"NVIDIA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Chelsea","family":"Finn","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ankur","family":"Handa","sequence":"additional","affiliation":[{"name":"NVIDIA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Tsung-Yi","family":"Lin","sequence":"additional","affiliation":[{"name":"NVIDIA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Gordon","family":"Wetzstein","sequence":"additional","affiliation":[{"name":"Stanford University"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Ming-Yu","family":"Liu","sequence":"additional","affiliation":[{"name":"NVIDIA"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Donglai","family":"Xiang","sequence":"additional","affiliation":[{"name":"NVIDIA"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73116-7_18"},{"key":"ref2","article-title":"Zero-shot robotic manipulation with pretrained image-editing diffusion models","author":"Black","year":"2023"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2023.xix.025"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00951"},{"key":"ref5","article-title":"Gr-2: A generative video-language-action model with web-scale knowledge for robot manipulation","author":"Cheang","year":"2024"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01370"},{"key":"ref7","article-title":"Deep compression autoencoder for efficient high-resolution diffusion models","author":"Chen","year":"2024"},{"key":"ref8","article-title":"Berkeley UR5 demonstration dataset","author":"Chen"},{"key":"ref9","article-title":"Pali-3 vision language models: Smaller, faster, stronger","author":"Chen","year":"2023"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1177\/02783649241273668"},{"key":"ref11","article-title":"Goal-conditioned imitation learning","volume":"32","author":"Ding","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"ref12","article-title":"Palm-e: An embodied multimodal language model","author":"Driess","year":"2023"},{"key":"ref13","article-title":"Vision-language models as success detectors","author":"Du","year":"2023"},{"key":"ref14","article-title":"Video language planning","author":"Du","year":"2023"},{"key":"ref15","article-title":"Learning universal policies via text-guided video generation","volume":"36","author":"Du","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2022.xviii.063"},{"key":"ref17","article-title":"Humanplus: Humanoid shadowing and imitation from humans","author":"Fu","year":"2024"},{"key":"ref18","article-title":"Mobile aloha: Learning bimanual mobile manipulation with low-cost whole-body teleoperation","volume-title":"Conference on Robot Learning (CoRL)","author":"Fu"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02219"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.622"},{"key":"ref21","first-page":"3766","article-title":"Scaling up and distilling down: Language-guided robot skill acquisition","volume-title":"Conference on Robot Learning","author":"Ha"},{"key":"ref22","article-title":"Visual chain-of-thought diffusion models","author":"Harvey","year":"2023"},{"key":"ref23","article-title":"Gaia-1: A generative world model for autonomous driving","author":"Hu","year":"2023"},{"key":"ref24","article-title":"Visual sketchpad: Sketching as a visual chain of thought for multi-modal language models","author":"Hu","year":"2024"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10160969"},{"key":"ref26","article-title":"Rekep: Spatio-temporal reasoning of relational keypoint constraints for robotic manipulation","author":"Huang","year":"2024"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/SmartWorld-UIC-ATC-SCALCOM-IOP-SCI.2019.00185"},{"key":"ref28","article-title":"Prismatic vlms: Investigating the design space of visually-conditioned language models","author":"Karamcheti","year":"2024"},{"key":"ref29","article-title":"Openvla: An open-source vision-language-action model","author":"Kim","year":"2024"},{"key":"ref30","article-title":"Videopoet: A large language model for zero-shot video generation","author":"Kondratyuk","year":"2023"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1145\/3600006.3613165"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01123"},{"key":"ref33","first-page":"19274","article-title":"Fast inference from transformers via speculative decoding","volume-title":"International Conference on Machine Learning","author":"Leviathan"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01416"},{"key":"ref35","article-title":"Dreamitate: Real-world visuomotor policy learning via video generation","author":"Liang","year":"2024"},{"key":"ref36","article-title":"Data scaling laws in imitation learning for robotic manipulation","author":"Lin","year":"2024"},{"key":"ref37","article-title":"Libero: Benchmarking knowledge transfer for lifelong robot learning","author":"Liu","year":"2023"},{"key":"ref38","article-title":"Visual instruction tuning","volume":"36","author":"Liu","year":"2024","journal-title":"Advances in neural information processing systems"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02497"},{"key":"ref40","article-title":"Eureka: Human-level reward design via coding large language models","author":"Ma","year":"2023"},{"key":"ref41","first-page":"655","article-title":"Where are we in the search for an artificial visual cortex for embodied intelligence?","volume":"36","author":"Majumdar","year":"2023","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref42","first-page":"879","article-title":"Roboturk: A crowdsourcing platform for robotic skill learning through imitation","volume-title":"Conference on Robot Learning","author":"Mandlekar"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10160396"},{"key":"ref44","article-title":"Robotic control via embodied chain-of-thought reasoning","author":"Micha\u0142","year":"2024"},{"key":"ref45","article-title":"Embodiedgpt: Vision-language pre-training via embodied chain of thought","volume":"36","author":"Mu","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref46","article-title":"Visual reinforcement learning with imagined goals","volume":"31","author":"Nair","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01327"},{"key":"ref48","article-title":"Open x-embodiment: Robotic learning datasets and rt-x models","author":"O\u2019Neill","year":"2023"},{"key":"ref49","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"International conference on machine learning","author":"Radford"},{"key":"ref50","article-title":"Visual chain of thought: bridging logical gaps with multimodal infillings","author":"Rose","year":"2023"},{"key":"ref51","first-page":"1838","article-title":"Latent plans for task-agnostic offline reinforcement learning","volume-title":"Conference on Robot Learning","author":"Rosete-Beas"},{"key":"ref52","article-title":"Distilbert, a distilled version of bert: smaller, faster, cheaper and lighter","author":"Sanh","year":"2019"},{"key":"ref53","article-title":"Visual cot: Unleashing chain-of-thought reasoning in multi-modal language models","author":"Shao","year":"2024"},{"key":"ref54","first-page":"894","article-title":"Cliport: What and where pathways for robotic manipulation","volume-title":"Conference on robot learning","author":"Shridhar"},{"key":"ref55","article-title":"Generative image as action models","author":"Shridhar","year":"2024"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10161317"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1145\/960126.806879"},{"key":"ref58","article-title":"Chameleon: Mixed-modal early-fusion foundation models","author":"Team","year":"2024"},{"key":"ref59","article-title":"Octo: An open-source generalist robot policy","author":"Team","year":"2024"},{"key":"ref60","first-page":"1723","article-title":"Bridgedata v2: A dataset for robot learning at scale","volume-title":"Conference on Robot Learning","author":"Walke"},{"key":"ref61","article-title":"Emu3: Next-token prediction is all you need","author":"Wang","year":"2024"},{"key":"ref62","article-title":"Chain-of-thought prompting elicits reasoning in large language models","author":"Wei","year":"2022","journal-title":"Advances in neural information processing systems"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2024.xx.092"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/lra.2025.3544909"},{"key":"ref65","article-title":"Janus: Decoupling visual encoding for unified multimodal understanding and generation","author":"Wu","year":"2024"},{"key":"ref66","article-title":"Unleashing large-scale video generative pre-training for visual robot manipulation","author":"Wu","year":"2023"},{"key":"ref67","article-title":"Vila-u: a unified foundation model integrating visual understanding and generation","author":"Wu","year":"2024"},{"key":"ref68","article-title":"Pandora: Towards general world model with natural language actions and video states","author":"Xiang","year":"2024"},{"key":"ref69","article-title":"Show-o: One single transformer to unify multimodal understanding and generation","author":"Xie","year":"2024"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2024.XX.093"},{"key":"ref71","article-title":"Learning interactive real-world simulators","author":"Yang","year":"2023"},{"key":"ref72","article-title":"Video as the new language for real-world decision making","author":"Yang","year":"2024"},{"key":"ref73","article-title":"An image is worth 32 tokens for reconstruction and generation","author":"Yu","year":"2024"},{"key":"ref74","article-title":"Language to rewards for robotic skill synthesis","author":"Yu","year":"2023"},{"key":"ref75","first-page":"15476","article-title":"Star: Bootstrapping reasoning with reasoning","volume":"35","author":"Zelikman","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref76","article-title":"Learning manipulation skills through robot chain-of-thought with sparse failure guidance","author":"Zhang","year":"2024"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2023.XIX.016"},{"key":"ref78","article-title":"3d-vla: A 3d vision-language-action generative world model","author":"Zhen","year":"2024"},{"key":"ref79","article-title":"Transfusion: Predict the next token and diffuse images with one multi-modal model","author":"Zhou","year":"2024"},{"key":"ref80","article-title":"Mine-dreamer: Learning to follow instructions via chain-of-imagination for simulated-world control","author":"Zhou","year":"2024"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10160594"},{"key":"ref82","article-title":"Fanuc manipulation: A dataset for learning-based manipulation with fanuc mate 200id robot","author":"Zhu","year":"2023"},{"key":"ref83","first-page":"1199","article-title":"Viola: Imitation learning for vision-based manipulation with object proposal priors","volume-title":"Conference on Robot Learning","author":"Zhu"}],"event":{"name":"2025 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)","location":"Nashville, TN, USA","start":{"date-parts":[[2025,6,10]]},"end":{"date-parts":[[2025,6,17]]}},"container-title":["2025 IEEE\/CVF Conference on Computer Vision and Pattern Recognition (CVPR)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11091818\/11091608\/11093669.pdf?arnumber=11093669","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,8,15]],"date-time":"2025-08-15T04:49:06Z","timestamp":1755233346000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11093669\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,6,10]]},"references-count":83,"URL":"https:\/\/doi.org\/10.1109\/cvpr52734.2025.00166","relation":{},"subject":[],"published":{"date-parts":[[2025,6,10]]}}}