{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,8,5]],"date-time":"2026-08-05T10:44:40Z","timestamp":1785926680874,"version":"3.56.0"},"reference-count":61,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"12","license":[{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,12,1]],"date-time":"2025-12-01T00:00:00Z","timestamp":1764547200000},"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":["IEEE Robot. Autom. Lett."],"published-print":{"date-parts":[[2025,12]]},"DOI":"10.1109\/lra.2025.3623027","type":"journal-article","created":{"date-parts":[[2025,10,17]],"date-time":"2025-10-17T17:42:36Z","timestamp":1760722956000},"page":"12955-12962","source":"Crossref","is-referenced-by-count":1,"title":["Don\u2019t Let Your Robot Be Harmful: Responsible Robotic Manipulation via Safety-As-Policy"],"prefix":"10.1109","volume":"10","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-6483-0650","authenticated-orcid":false,"given":"Minheng","family":"Ni","sequence":"first","affiliation":[{"name":"Hong Kong Polytechnic University, Kowloon, Hong Kong, SAR, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6302-8348","authenticated-orcid":false,"given":"Lei","family":"Zhang","sequence":"additional","affiliation":[{"name":"University of Hamburg, Hamburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8672-5816","authenticated-orcid":false,"given":"Zihan","family":"Chen","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Harbin, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-8579-0547","authenticated-orcid":false,"given":"Kaixin","family":"Bai","sequence":"additional","affiliation":[{"name":"University of Hamburg, Hamburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0375-9146","authenticated-orcid":false,"given":"Zhaopeng","family":"Chen","sequence":"additional","affiliation":[{"name":"Agile Robots, Munich, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7856-5760","authenticated-orcid":false,"given":"Jianwei","family":"Zhang","sequence":"additional","affiliation":[{"name":"University of Hamburg, Hamburg, Germany"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-2078-4215","authenticated-orcid":false,"given":"Lei","family":"Zhang","sequence":"additional","affiliation":[{"name":"Hong Kong Polytechnic University, Kowloon, Hong Kong, SAR, China"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3330-783X","authenticated-orcid":false,"given":"Wangmeng","family":"Zuo","sequence":"additional","affiliation":[{"name":"Harbin Institute of Technology, Harbin, China"}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/JAS.2017.7510604"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.17973\/MMSJ.2016_06_201611"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/ACCESS.2023.3310935"},{"key":"ref4","first-page":"69900","article-title":"STEVE-1: A generative model for text-to-behavior in minecraft","volume-title":"Proc. 37th Adv. Conf. Neural Inf. Process. Syst.","author":"Lifshitz","year":"2024"},{"key":"ref5","article-title":"Steve series: Step-by-step construction of agent systems in minecraft","author":"Zhao","year":"2024"},{"key":"ref6","article-title":"Contactdexnet: Multi-fingered robotic hand grasping in cluttered environments through hand-object contact semantic mapping","author":"Zhang","year":"2024"},{"key":"ref7","article-title":"M4diffuser: Multi-view diffusion policy with manipulability-aware control for robust mobile manipulation","author":"Dong","year":"2025"},{"key":"ref8","article-title":"Funcanon: Learning pose-aware action primitives via functional object canonicalization for generalizable robotic manipulation","author":"Xu","year":"2025"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/IROS51168.2021.9636342"},{"key":"ref10","article-title":"Creative robot tool use with large language models","author":"Xu","year":"2023"},{"key":"ref11","article-title":"Generalizable long-horizon manipulations with large language models","author":"Zhou","year":"2023"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2023.xix.025"},{"key":"ref13","article-title":"Language to rewards for robotic skill synthesis","author":"Yu","year":"2023"},{"key":"ref14","article-title":"Eureka: Human-level reward design via coding large language models","author":"Ma","year":"2023"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-control-060117-104848"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1126\/science.aat8414"},{"key":"ref17","article-title":"R3m: A universal visual representation for robot manipulation","author":"Nair","year":"2022"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.52202\/079017-3188"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.4172\/2168-9695.1000139"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1016\/j.inffus.2019.12.012"},{"key":"ref21","first-page":"80079","article-title":"Jailbroken: How does LLM safety training fail?","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"36","author":"Wei","year":"2024"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02157"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01042"},{"issue":"2","key":"ref24","first-page":"275","article-title":"Advancing nsfw detection in AI: Training models to detect drawings, animations, and assess degrees of sexiness","volume":"2","author":"Guzman","year":"2023","journal-title":"J. Knowl. Learn. Sci. Technol."},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/ICSE.2012.6227177"},{"key":"ref26","article-title":"GPT-4 technical report","author":"Achiam","year":"2023"},{"key":"ref27","article-title":"Gemini: A family of highly capable multimodal models","author":"Team","year":"2023"},{"key":"ref28","article-title":"The LLAMA 3 herd of models","author":"Dubey","year":"2024"},{"key":"ref29","article-title":"Training a helpful and harmless assistant with reinforcement learning from human feedback","author":"Bai","year":"2022"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/SP40001.2021.00019"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1038\/s42256-025-00985-0"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2024.findings-acl.107"},{"key":"ref33","article-title":"In-context unlearning: Language models as few shot unlearners","author":"Pawelczyk","year":"2023"},{"key":"ref34","article-title":"Injecting bias in text-to-image models via composite-trigger backdoors","author":"Naseh","year":"2024"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01141"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-73661-2_20"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i19.30144"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72670-5_18"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/IROS45743.2020.9341522"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1146\/annurev-control-071723-102940"},{"key":"ref41","first-page":"25621","article-title":"Learning barrier certificates: Towards safe reinforcement learning with zero training-time violations","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Luo","year":"2021"},{"key":"ref42","first-page":"1877","article-title":"Learning to be safe: Deep RL with a safety critic","volume":"33","author":"Srinivasan","year":"2020","journal-title":"Adv. Neural Inf. Process. Syst."},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2024.3396368"},{"key":"ref44","first-page":"1554","article-title":"Safe driving via expert guided policy optimization","volume-title":"Proc. Conf. Robot Learn.","author":"Peng","year":"2022"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.15607\/rss.2025.xxi.010"},{"key":"ref46","article-title":"Openvla: An open-source vision-language-action model","author":"Kim","year":"2024"},{"key":"ref47","article-title":"RT-2: Vision-language-action models transfer web knowledge to robotic control","author":"Brohan","year":"2023"},{"key":"ref48","article-title":"Vision-language foundation models as effective robot imitators","author":"Li","year":"2023"},{"key":"ref49","first-page":"287","article-title":"Do as i can, not as i say: Grounding language in robotic affordances","volume-title":"Proc. Conf. Robot Learn.","author":"Brohan","year":"2023"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/ICRA48891.2023.10160591"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1007\/s10514-023-10139-z"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1007\/s10514-023-10131-7"},{"key":"ref53","first-page":"540","article-title":"Voxposer: Composable 3D value maps for robotic manipulation with language models","volume-title":"Proc. 7th Annu. Conf. Robot Learn.","author":"Huang","year":"2023"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/LRA.2024.3477090"},{"key":"ref55","article-title":"What foundation models can bring for robot learning in manipulation: A survey","author":"Li","year":"2024"},{"key":"ref56","article-title":"Language-conditioned learning for robotic manipulation: A survey","author":"Zhou","year":"2023"},{"key":"ref57","article-title":"Gpt-4o system card","author":"Hurst","year":"2024"},{"key":"ref58","first-page":"8821","article-title":"Zero-shot text-to-image generation","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Ramesh","year":"2021"},{"key":"ref59","article-title":"Language models are few-shot learners","author":"Brown","year":"2020"},{"key":"ref60","article-title":"Visual-o1: Understanding ambiguous instructions via multi-modal multi-turn chain-of-thoughts reasoning","author":"Ni","year":"2024"},{"key":"ref61","first-page":"24824","article-title":"Chain-of-thought prompting elicits reasoning in large language models","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"35","author":"Wei","year":"2022"}],"container-title":["IEEE Robotics and Automation Letters"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/7083369\/11215960\/11206481.pdf?arnumber=11206481","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,11,24]],"date-time":"2025-11-24T19:02:59Z","timestamp":1764010979000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11206481\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,12]]},"references-count":61,"journal-issue":{"issue":"12"},"URL":"https:\/\/doi.org\/10.1109\/lra.2025.3623027","relation":{},"ISSN":["2377-3766","2377-3774"],"issn-type":[{"value":"2377-3766","type":"electronic"},{"value":"2377-3774","type":"electronic"}],"subject":[],"published":{"date-parts":[[2025,12]]}}}