{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,9,25]],"date-time":"2026-09-25T07:02:37Z","timestamp":1790319757121,"version":"4.1.0"},"reference-count":82,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"8","license":[{"start":{"date-parts":[[2026,8,1]],"date-time":"2026-08-01T00:00:00Z","timestamp":1785542400000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/creativecommons.org\/licenses\/by\/4.0\/legalcode"}],"funder":[{"name":"MTR Research Funding Scheme","award":["CHU-24003"],"award-info":[{"award-number":["CHU-24003"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Circuits Syst. Video Technol."],"published-print":{"date-parts":[[2026,8]]},"DOI":"10.1109\/tcsvt.2026.3694713","type":"journal-article","created":{"date-parts":[[2026,5,19]],"date-time":"2026-05-19T19:51:24Z","timestamp":1779220284000},"page":"11027-11041","source":"Crossref","is-referenced-by-count":4,"title":["MotionGPT-2: A General-Purpose Motion-Language Model for Motion Generation and Understanding"],"prefix":"10.1109","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0009-0006-6796-472X","authenticated-orcid":false,"given":"Yuan","family":"Wang","sequence":"first","affiliation":[{"name":"Tsinghua University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0009-8712-8747","authenticated-orcid":false,"given":"Di","family":"Huang","sequence":"additional","affiliation":[{"name":"The University of Sydney","place":["Sydney, NSW, Australia"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"given":"Yaqi","family":"Zhang","sequence":"additional","affiliation":[{"name":"University of Science and Technology of China","place":["Hefei, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-9163-2761","authenticated-orcid":false,"given":"Wanli","family":"Ouyang","sequence":"additional","affiliation":[{"name":"Shanghai Artificial Intelligence Laboratory","place":["Shanghai, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2644-717X","authenticated-orcid":false,"given":"Jile","family":"Jiao","sequence":"additional","affiliation":[{"name":"Tsinghua University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1309-8954","authenticated-orcid":false,"given":"Xuetao","family":"Feng","sequence":"additional","affiliation":[{"name":"Tsinghua University","place":["Beijing, China"]}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0136-9603","authenticated-orcid":false,"given":"Dan","family":"Xu","sequence":"additional","affiliation":[{"name":"HKUST"}],"role":[{"vocabulary":"crossref","role":"author"}]},{"ORCID":"https:\/\/orcid.org\/0009-0005-0067-339X","authenticated-orcid":false,"given":"Shixiang","family":"Tang","sequence":"additional","affiliation":[{"name":"The Chinese University of Hong Kong","place":["Shatin, Hong Kong"]}],"role":[{"vocabulary":"crossref","role":"author"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v38i7.28567"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2022.3163782"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3255186"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3284013"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2022.3232330"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2021.3057992"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3318557"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3298755"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2019.2953753"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2025.3531390"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3268062"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2021.3064035"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3274228"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2025.3532495"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3244786"},{"key":"ref16","article-title":"MotionDiffuse: Text-driven human motion generation with diffusion model","author":"Zhang","year":"2022","journal-title":"arXiv:2208.15001"},{"key":"ref17","first-page":"1","article-title":"Human motion diffusion model","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Tevet"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1145\/3485664"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1145\/3641519.3657414"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00205"},{"key":"ref21","article-title":"OmniControl: Control any joint at any time for human motion generation","author":"Xie","year":"2023","journal-title":"arXiv:2310.08580"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01880"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01415"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01726"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00186"},{"key":"ref26","first-page":"20067","article-title":"MotionGPT: Human motion as a foreign language","volume-title":"Proc. Adv. Neural Inf. Process. Syst. 36","author":"Chen"},{"key":"ref27","article-title":"M3GPT: An advanced multimodal, multitask framework for motion comprehension and generation","author":"Luo","year":"2024","journal-title":"arXiv:2405.16273"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00040"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00509"},{"key":"ref30","article-title":"MiniGPT-4: Enhancing vision-language understanding with advanced large language models","author":"Zhu","year":"2023","journal-title":"arXiv:2304.10592"},{"key":"ref31","article-title":"Vision-language models as success detectors","author":"Du","year":"2023","journal-title":"arXiv:2303.07280"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.52202\/075280-1516"},{"key":"ref33","article-title":"MPLUG-owl: Modularization empowers large language models with multimodality","author":"Ye","year":"2023","journal-title":"arXiv:2304.14178"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1007\/s11432-024-4321-9"},{"key":"ref35","article-title":"LoRA: Low-rank adaptation of large language models","author":"Hu","year":"2021","journal-title":"arXiv:2106.09685"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00385"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1089\/big.2016.0028"},{"key":"ref38","first-page":"25268","article-title":"Motion-X: A large-scale 3D expressive whole-body human motion dataset","volume-title":"Proc. Adv. Neural Inf. Process. Syst. 36","author":"Cai"},{"key":"ref39","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018","journal-title":"arXiv:1810.04805"},{"key":"ref40","article-title":"Improving language understanding by generative pre-training","author":"Radford","year":"2018"},{"key":"ref41","article-title":"Language models are unsupervised multitask learners","author":"Radford","year":"2019"},{"key":"ref42","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. NIPS","author":"Brown"},{"key":"ref43","article-title":"Gpt-4 technical report","volume-title":"arXiv:2303.08774","author":"Achiam","year":"2023"},{"key":"ref44","article-title":"LLaMA: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv:2302.13971"},{"issue":"1","key":"ref45","first-page":"5485","article-title":"Exploring the limits of transfer learning with a unified text-to-text transformer","volume":"21","author":"Raffel","year":"2020","journal-title":"J. Mach. Learn. Res."},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.emnlp-main.243"},{"key":"ref47","article-title":"P-Tuning v2: Prompt tuning can be comparable to fine-tuning universally across scales and tasks","author":"Liu","year":"2021","journal-title":"arXiv:2110.07602"},{"key":"ref48","article-title":"Knowledgeable prompt-tuning: Incorporating knowledge into prompt verbalizer for text classification","author":"Hu","year":"2021","journal-title":"arXiv:2108.02035"},{"key":"ref49","first-page":"2790","article-title":"Parameter-efficient transfer learning for NLP","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Houlsby"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-long.172"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2021.acl-short.103"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-20047-2_21"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.5244\/C.31.119"},{"key":"ref54","article-title":"Auto-conditioned recurrent networks for extended complex human motion synthesis","author":"Li","year":"2017","journal-title":"arXiv:1707.05363"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1145\/3394171.3413635"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.01315"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/CVPRW.2018.00191"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.497"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.01080"},{"key":"ref60","article-title":"TrajeVAE: Controllable human motion generation from trajectories","author":"Kania","year":"2021","journal-title":"arXiv:2104.00351"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00143"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19833-5_34"},{"key":"ref63","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","volume-title":"Proc. Int. Conf. Mach. Learn.","volume":"139","author":"Radford"},{"key":"ref64","article-title":"MotionLLM: Multimodal motion-language learning with large language models","author":"Wu","year":"2024","journal-title":"arXiv:2405.17013"},{"key":"ref65","first-page":"6306","article-title":"Neural discrete representation learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"Van Den Oord"},{"key":"ref66","article-title":"HumanTOMATO: Text-aligned whole-body motion generation","author":"Lu","year":"2023","journal-title":"arXiv:2310.12978"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2024.3355414"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.00053"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.52202\/075280-0681"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00554"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/ICAR.2015.7251476"},{"key":"ref72","volume-title":"CMU Graphics Lab Motion Capture Database","year":"2000"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01123"},{"key":"ref74","article-title":"Decoupled weight decay regularization","author":"Loshchilov","year":"2017","journal-title":"arXiv:1711.05101"},{"key":"ref75","article-title":"Qwen2.5-VL technical report","volume-title":"arXiv:2502.13923","author":"Bai","year":"2025"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.3115\/1073083.1073135"},{"key":"ref77","article-title":"BERTScore: Evaluating text generation with bert","author":"Zhang","year":"2019","journal-title":"arXiv:1904.09675"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00799"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58545-7_20"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00338"},{"key":"ref81","article-title":"DiverseMotion: Towards diverse human motion generation via discrete diffusion","author":"Lou","year":"2023","journal-title":"arXiv:2309.01372"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1631\/FITEE.2100463"}],"container-title":["IEEE Transactions on Circuits and Systems for Video Technology"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/76\/11643439\/11527021.pdf?arnumber=11527021","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,8,6]],"date-time":"2026-08-06T19:08:16Z","timestamp":1786043296000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11527021\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,8]]},"references-count":82,"journal-issue":{"issue":"8"},"URL":"https:\/\/doi.org\/10.1109\/tcsvt.2026.3694713","relation":{},"ISSN":["1051-8215","1558-2205"],"issn-type":[{"value":"1051-8215","type":"print"},{"value":"1558-2205","type":"electronic"}],"subject":[],"published":{"date-parts":[[2026,8]]}}}