{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,4,17]],"date-time":"2026-04-17T22:40:11Z","timestamp":1776465611418,"version":"3.51.2"},"reference-count":53,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,5,19]],"date-time":"2025-05-19T00:00:00Z","timestamp":1747612800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,5,19]],"date-time":"2025-05-19T00:00:00Z","timestamp":1747612800000},"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":[[2025,5,19]]},"DOI":"10.1109\/icra55743.2025.11127829","type":"proceedings-article","created":{"date-parts":[[2025,9,2]],"date-time":"2025-09-02T17:28:56Z","timestamp":1756834136000},"page":"5930-5936","source":"Crossref","is-referenced-by-count":1,"title":["Robotic-CLIP: Fine-Tuning CLIP on Action Data for Robotic Applications"],"prefix":"10.1109","author":[{"given":"Nghia","family":"Nguyen","sequence":"first","affiliation":[{"name":"FPT Software AI Center,Vietnam"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Minh N.","family":"Vu","sequence":"additional","affiliation":[{"name":"Automation & Control Institute (ACIN), TU Wien,Vienna,Austria"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Tung D.","family":"Ta","sequence":"additional","affiliation":[{"name":"The University of Tokyo,Japan"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Baoru","family":"Huang","sequence":"additional","affiliation":[{"name":"University of Liverpool,Department of Computer Science,UK"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Thieu","family":"Vo","sequence":"additional","affiliation":[{"name":"National University of Singapore,Singapore"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ngan","family":"Le","sequence":"additional","affiliation":[{"name":"University of Arkansas,USA"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Anh","family":"Nguyen","sequence":"additional","affiliation":[{"name":"University of Liverpool,Department of Computer Science,UK"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","author":"Brohan","year":"2023","journal-title":"Rt-2: Vision-language-action models transfer web knowledge to robotic control"},{"key":"ref2","article-title":"Lm-nav: Robotic navigation with large pre-trained models of language, vision, and action","author":"Shah","year":"2023","journal-title":"CoRL"},{"key":"ref3","article-title":"Real-world robot learning with masked visual pre-training","author":"Radosavovic","year":"2023","journal-title":"CoRL"},{"key":"ref4","article-title":"Liv: Language-image repre-sentations and rewards for robotic control","volume-title":"ICML","author":"Ma","year":"2023"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00678"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01695"},{"key":"ref7","article-title":"Cliport: What and where pathways for robotic manipulation","author":"Shridhar","year":"2022","journal-title":"CoRL"},{"key":"ref8","article-title":"Octo: An open-source generalist robot policy","author":"Team","year":"2024","journal-title":"arXiv preprint"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01370"},{"key":"ref10","author":"Huang","year":"2022","journal-title":"Inner mono-logue: Embodied reasoning through planning with language models"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA57147.2024.10610216"},{"key":"ref12","article-title":"Learning transferable visual models from natural language supervision","volume-title":"ICML","author":"Radford","year":"2021"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01237"},{"key":"ref14","article-title":"Blip-2: Bootstrapping language-image pre-training with frozen image encoders and large language models","volume-title":"ICML","author":"Li","year":"2023"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2023.XIX.032"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01366"},{"key":"ref17","author":"Dosovitskiy","year":"2020","journal-title":"An image is worth 16x16 words: Transformers for image recognition at scale"},{"key":"ref18","article-title":"Bert: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018","journal-title":"arXiv preprint"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA57147.2024.10611277"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.emnlp-main.544"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10160969"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/ICCVW60793.2023.00218"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00836"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72970-6_3"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00371"},{"key":"ref26","article-title":"Flair: An easy-to-use framework for state-of-the-art nlp","volume-title":"NAACL","author":"Akbik","year":"2019"},{"key":"ref27","article-title":"Roboclip: One demonstration is enough to learn robot policies","author":"Sontakke","year":"2023","journal-title":"NeurIPS"},{"key":"ref28","author":"Nair","year":"2022","journal-title":"R3m: A universal visual representation for robot manipulation"},{"key":"ref29","article-title":"Vip: Towards universal visual reward and representation via value-implicit pre-training","volume-title":"ICLR","author":"Ma","year":"2023"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/IROS45743.2020.9340777"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48506.2021.9561398"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10161041"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2018.XIV.021"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/IROS55552.2023.10341553"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA57147.2024.10610247"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/WACV45572.2020.9093545"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01058"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00289"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP43922.2022.9747631"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10160429"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1109\/IROS47612.2022.9981055"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA57147.2024.10610095"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01934"},{"key":"ref44","article-title":"The","volume-title":"ICCV","author":"Goyal","year":"2017"},{"key":"ref45","article-title":"Bridgedata v2: A dataset for robot learning at scale","author":"Walke","year":"2023","journal-title":"CoRL"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/icra57147.2024.10611293"},{"key":"ref47","article-title":"Videocomposer: Compositional video synthesis with motion controllability","volume":"36","author":"Wang","year":"2024","journal-title":"NeurIPS"},{"key":"ref48","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-023-01891-x"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1109\/IROS58592.2024.10802007"},{"key":"ref50","article-title":"Learning language-conditioned robot behavior from offline data and crowd-sourced annotation","author":"Nair","year":"2022","journal-title":"CoRL"},{"key":"ref51","author":"Gupta","year":"2019","journal-title":"Relay policy learning: Solving long-horizon tasks via imitation and reinforcement learning"},{"key":"ref52","article-title":"Carla: An open urban driving simulator","author":"Dosovitskiy","year":"2017","journal-title":"CoRL"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.1016\/j.mechatronics.2023.102970"}],"event":{"name":"2025 IEEE International Conference on Robotics and Automation (ICRA)","location":"Atlanta, GA, USA","start":{"date-parts":[[2025,5,19]]},"end":{"date-parts":[[2025,5,23]]}},"container-title":["2025 IEEE International Conference on Robotics and Automation (ICRA)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11127273\/11127223\/11127829.pdf?arnumber=11127829","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,3]],"date-time":"2025-09-03T06:46:09Z","timestamp":1756881969000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11127829\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,19]]},"references-count":53,"URL":"https:\/\/doi.org\/10.1109\/icra55743.2025.11127829","relation":{},"subject":[],"published":{"date-parts":[[2025,5,19]]}}}