{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,9]],"date-time":"2026-05-09T17:34:38Z","timestamp":1778348078802,"version":"3.51.4"},"reference-count":69,"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\/icra55743.2025.11127531","type":"proceedings-article","created":{"date-parts":[[2025,9,2]],"date-time":"2025-09-02T17:28:56Z","timestamp":1756834136000},"page":"4182-4189","source":"Crossref","is-referenced-by-count":1,"title":["Robo-MUTUAL: Robotic Multimodal Task Specification via Unimodal Learning"],"prefix":"10.1109","author":[{"given":"Jianxiong","family":"Li","sequence":"first","affiliation":[{"name":"Institute for AI Industry Research (AIR), Tsinghua University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhihao","family":"Wang","sequence":"additional","affiliation":[{"name":"College of Engineering, Peking University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jinliang","family":"Zheng","sequence":"additional","affiliation":[{"name":"Institute for AI Industry Research (AIR), Tsinghua University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xiaoai","family":"Zhou","sequence":"additional","affiliation":[{"name":"University of Toronto"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Guanming","family":"Wang","sequence":"additional","affiliation":[{"name":"University College London"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Guanglu","family":"Song","sequence":"additional","affiliation":[{"name":"Sensetime Research"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Yu","family":"Liu","sequence":"additional","affiliation":[{"name":"Sensetime Research"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Jingjing","family":"Liu","sequence":"additional","affiliation":[{"name":"Institute for AI Industry Research (AIR), Tsinghua University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Ya-Qin","family":"Zhang","sequence":"additional","affiliation":[{"name":"Institute for AI Industry Research (AIR), Tsinghua University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Junzhi","family":"Yu","sequence":"additional","affiliation":[{"name":"College of Engineering, Peking University"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xianyuan","family":"Zhan","sequence":"additional","affiliation":[{"name":"Institute for AI Industry Research (AIR), Tsinghua University"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"Mutex: Learning unified policies from multimodal task specifications","volume-title":"7th Annual Conference on Robot Learning","author":"Shah","year":"2023"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2024.XX.121"},{"key":"ref3","article-title":"Octo: An open-source generalist robot policy","volume-title":"Proceedings of Robotics: Science and Systems, (Delft, Netherlands)","author":"Model Team","year":"2024"},{"key":"ref4","article-title":"Vima: General robot manipulation with multimodal prompts","volume-title":"Fortieth International Conference on Machine Learning","author":"Jiang","year":"2023"},{"key":"ref5","article-title":"Palm-e: An embodied multimodal language model","author":"Driess","year":"2023","journal-title":"arXiv preprint"},{"key":"ref6","article-title":"Goal representations for instruction following: A semi-supervised language interface to control","volume-title":"Conference on Robot Learning, pp. 38943908, PMLR","author":"Myers","year":"2023"},{"key":"ref7","article-title":"Using both demonstrations and language instructions to efficiently learn robotic tasks","volume-title":"The Eleventh International Conference on Learning Representations","author":"Yu","year":"2023"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW53098.2021.00177"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2023.xix.029"},{"key":"ref10","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"in International conference on machine learning","author":"Radford"},{"key":"ref11","article-title":"Decisionnce: Embodied multimodal representations via implicit preference learning","volume-title":"Forty-first International Conference on Machine Learning","author":"Li","year":"2024"},{"key":"ref12","article-title":"Connect, collapse, corrupt:Learning cross-modal tasks with uni-modal data","author":"Zhang","year":"2024","journal-title":"arXiv preprint"},{"key":"ref13","first-page":"17612","article-title":"Mind the gap: Understanding the modality gap in multi-modal contrastive representation learning","volume":"35","author":"Liang","year":"2022","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref14","article-title":"The platonic representation hypothesis","volume-title":"arXiv preprint","author":"Huh","year":"2024"},{"key":"ref15","article-title":"Decap: Decoding clip latents for zero-shot captioning via text-only training","author":"Li","year":"2023","journal-title":"arXiv preprint"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.622"},{"key":"ref17","first-page":"720","article-title":"Scaling egocentric vision: The epic-kitchens dataset","volume-title":"in Proceedings of the European conference on computer vision (ECCV)","author":"Damen","year":"2018"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01842"},{"key":"ref19","first-page":"23301","article-title":"Liv: Language-image representations and rewards for robotic control","volume-title":"in International Conference on Machine Learning","author":"Ma"},{"key":"ref20","article-title":"R3m: A universal visual representation for robot manipulation","author":"Nair","year":"2022","journal-title":"arXiv preprint"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2023.XIX.032"},{"key":"ref22","article-title":"Open x-embodiment: Robotic learning datasets and rt-x models","author":"Padalkar","year":"2023","journal-title":"arXiv preprint"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2024.XX.120"},{"key":"ref24","article-title":"Libero: Benchmarking knowledge transfer for lifelong robot learning","volume":"36","author":"Liu","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2023.xix.025"},{"key":"ref26","article-title":"Rt-2: Vision-language-action models transfer web knowledge to robotic control","author":"Brohan","year":"2023","journal-title":"arXiv preprint"},{"key":"ref27","article-title":"Openvla: An open-source vision-language-action model","author":"Kim","year":"2024","journal-title":"arXiv preprint"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2023.xix.029"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2024.XX.049"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2024.XX.025"},{"key":"ref31","article-title":"VIP: Towards universal visual reward and representation via value-implicit pre-training","volume-title":"The Eleventh International Conference on Learning Representations","author":"Ma","year":"2023"},{"key":"ref32","article-title":"Planning with goalconditioned policies","volume":"32","author":"Nasiriany","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"ref33","first-page":"893","article-title":"Can foundation models perform zero-shot task specification for robot manipulation?","volume-title":"in Learning for dynamics and control conference","author":"Cui"},{"key":"ref34","article-title":"From play to policy: Conditional behavior generation from uncurated robot data","volume-title":"The Eleventh International Conference on Learning Representations","author":"Cui","year":"2023"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2018.xiv.002"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2020.2977835"},{"key":"ref37","first-page":"783","article-title":"Task-embedded control networks for few-shot imitation learning","volume-title":"in Conference on robot learning","author":"James"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2021.xvii.047"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2022.3196123"},{"key":"ref40","article-title":"Bc-z: Zero-shot task generalization with robotic imitation learning","volume-title":"Conference on Robot Learning, pp. 9911002, PMLR","author":"Jang","year":"2022"},{"key":"ref41","article-title":"Blip: Bootstrapping languageimage pre-training for unified vision-language understanding and generation","volume-title":"International conference on machine learning, pp. 1288812900, PMLR","author":"Li","year":"2022"},{"key":"ref42","first-page":"1973019742","article-title":"Blip-2: Bootstrapping languageimage pre-training with frozen image encoders and large language models","volume-title":"International conference on machine learning","author":"Li","year":"2023"},{"key":"ref43","article-title":"Clipcap: Clip prefix for image captioning","author":"Mokady","year":"2021","journal-title":"arXiv preprint"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.emnlp-main.595"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-024-02202-8"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01738"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01441"},{"key":"ref48","first-page":"13031315","article-title":"Learning language-conditioned robot behavior from offline data and crowdsourced annotation","volume-title":"Conference on Robot Learning","author":"Nair","year":"2022"},{"key":"ref49","article-title":"Bridgedata v2: A dataset for robot learning at scale","volume-title":"Conference on Robot Learning (CoRL)","author":"Walke","year":"2023"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1038\/s41467-024-50309-w"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1038\/s41598-020-58329-4"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1201\/9781439812174-23"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.findings-emnlp.299"},{"key":"ref54","article-title":"Two effects, one trigger: On the modality gap, object bias, and information imbalance in contrastive vision-language representation learning","volume-title":"ICLR 2024 Workshop on Mathematical and Empirical Understanding of Foundation Models","author":"Schrodi","year":"2024"},{"key":"ref55","article-title":"Dualdice: Behavior-agnostic estimation of discounted stationary distribution corrections","volume":"32","author":"Nachum","year":"2019","journal-title":"Advances in neural information processing systems"},{"key":"ref56","article-title":"Mind the gap: Offline policy optimization for imperfect rewards","volume-title":"The Eleventh International Conference on Learning Representations","author":"Li","year":"2023"},{"issue":"11","key":"ref57","article-title":"Visualizing data using t-sne","volume":"9","author":"Van der Maaten","year":"2008","journal-title":"Journal of machine learning research"},{"key":"ref58","article-title":"Understanding dimensional collapse in contrastive self-supervised learning","volume-title":"International Conference on Learning Representations","author":"Jing","year":"2022"},{"key":"ref59","article-title":"Diagnosing and rectifying vision models using language","volume-title":"International Conference on Learning Representations (ICLR)","author":"Zhang","year":"2023"},{"key":"ref60","article-title":"Arcsin: Adaptive ranged cosine similarity injected noise for language-driven visual tasks","author":"Liu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11671"},{"key":"ref63","article-title":"Idql: Implicit q-learning as an actor-critic method with diffusion policies","author":"Hansen-Estruch","year":"2023","journal-title":"arXiv preprint"},{"key":"ref64","first-page":"6840","article-title":"Denoising diffusion probabilistic models","volume":"33","author":"Ho","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1177\/02783649241273668"},{"key":"ref66","article-title":"Safe offline reinforcement learning with feasibility-guided diffusion model","author":"Zheng","year":"2024","journal-title":"arXiv preprint"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.15607\/RSS.2023.XIX.016"},{"key":"ref68","article-title":"Scaling robot policy learning via zero-shot labeling with foundation models","volume-title":"2nd Workshop on Mobile Manipulation and Embodied Intelligence at ICRA 2024","author":"Blank","year":"2024"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.emnlp-main.64"}],"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\/11127531.pdf?arnumber=11127531","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,9,3]],"date-time":"2025-09-03T06:05:08Z","timestamp":1756879508000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11127531\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,5,19]]},"references-count":69,"URL":"https:\/\/doi.org\/10.1109\/icra55743.2025.11127531","relation":{},"subject":[],"published":{"date-parts":[[2025,5,19]]}}}