{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,23]],"date-time":"2025-10-23T01:10:17Z","timestamp":1761181817052,"version":"build-2065373602"},"reference-count":40,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","license":[{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,1,1]],"date-time":"2025-01-01T00:00:00Z","timestamp":1735689600000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"name":"National Key R&D Program of China","award":["2024YFB3311602"],"award-info":[{"award-number":["2024YFB3311602"]}]},{"name":"Natural Science Foundation of Chinaunder","award":["62272144"],"award-info":[{"award-number":["62272144"]}]},{"name":"Anhui Natural Science Foundationunder","award":["2408085MF157"],"award-info":[{"award-number":["2408085MF157"]}]},{"name":"Major Project of Anhui Provinceunder","award":["2408085J040"],"award-info":[{"award-number":["2408085J040"]}]},{"name":"University Synergy Innovation Program of Anhui Provinceunder","award":["GXXT-2023-015"],"award-info":[{"award-number":["GXXT-2023-015"]}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Multimedia"],"published-print":{"date-parts":[[2025]]},"DOI":"10.1109\/tmm.2025.3599089","type":"journal-article","created":{"date-parts":[[2025,9,1]],"date-time":"2025-09-01T19:31:31Z","timestamp":1756755091000},"page":"7729-7743","source":"Crossref","is-referenced-by-count":0,"title":["Instructive Probabilistic Transformer for Complex Action Recognition"],"prefix":"10.1109","volume":"27","author":[{"ORCID":"https:\/\/orcid.org\/0000-0001-9834-4730","authenticated-orcid":false,"given":"Zhao","family":"Xie","sequence":"first","affiliation":[{"name":"Hefei University of Technology, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Longsheng","family":"Lu","sequence":"additional","affiliation":[{"name":"Hefei University of Technology, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-7332-5653","authenticated-orcid":false,"given":"Kewei","family":"Wu","sequence":"additional","affiliation":[{"name":"Hefei University of Technology, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Zhehan","family":"Kan","sequence":"additional","affiliation":[{"name":"Shenzhen International Graduate School, Tsinghua University, Shenzhen, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Xingming","family":"Yang","sequence":"additional","affiliation":[{"name":"Hefei University of Technology, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2594-254X","authenticated-orcid":false,"given":"Dan","family":"Guo","sequence":"additional","affiliation":[{"name":"Hefei University of Technology, Hefei, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1016\/j.cviu.2016.10.018"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3085755"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2022.3223688"},{"key":"ref4","first-page":"510","article-title":"Hollywood in homes: Crowdsourcing data collection for activity understanding","volume-title":"Proc. 14th Eur. Conf. Comput. Vis.","author":"Gunnar","year":"2016"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.105"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01942"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.502"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00034"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00630"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00813"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00037"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3302471"},{"article-title":"VideoGraph: Recognizing minutes-long human activities in videos","year":"2019","author":"Hussein","key":"ref13"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00887"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.01332"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00675"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00640"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00748"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02192"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2010.11929"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.337"},{"key":"ref22","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2022.3224327"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2024.3358415"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3664647.3688973"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1016\/j.patcog.2021.107925"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00618"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3312856"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2023.3333206"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2024.3361157"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1016\/j.ins.2020.11.026"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1049\/ipr2.12265"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/JSEN.2023.3294360"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/TIP.2023.3265261"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00154"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01352"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1145\/3712012"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19833-5_23"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01755"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3076172"},{"key":"ref40","first-page":"2579","article-title":"Visualizing data using t-SNE","volume":"9","author":"Maaten","year":"2008","journal-title":"J. Mach. Learn. Res."}],"container-title":["IEEE Transactions on Multimedia"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/6046\/10844992\/11145959.pdf?arnumber=11145959","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,22]],"date-time":"2025-10-22T17:25:27Z","timestamp":1761153927000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11145959\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025]]},"references-count":40,"URL":"https:\/\/doi.org\/10.1109\/tmm.2025.3599089","relation":{},"ISSN":["1520-9210","1941-0077"],"issn-type":[{"type":"print","value":"1520-9210"},{"type":"electronic","value":"1941-0077"}],"subject":[],"published":{"date-parts":[[2025]]}}}