{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,8]],"date-time":"2026-05-08T13:36:11Z","timestamp":1778247371563,"version":"3.51.4"},"reference-count":59,"publisher":"IEEE","license":[{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,19]],"date-time":"2025-10-19T00:00:00Z","timestamp":1760832000000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62176047,62476051"],"award-info":[{"award-number":["62176047,62476051"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100018542","name":"Sichuan Natural Science Foundation","doi-asserted-by":"publisher","award":["2024NSFTD0041"],"award-info":[{"award-number":["2024NSFTD0041"]}],"id":[{"id":"10.13039\/501100018542","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.01950","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"20981-20990","source":"Crossref","is-referenced-by-count":1,"title":["The Devil Is in the Spurious Correlations: Boosting Moment Retrieval With Dynamic Learning"],"prefix":"10.1109","author":[{"given":"Xinyang","family":"Zhou","sequence":"first","affiliation":[{"name":"University of Electronic Science and Technology of China,China"}]},{"given":"Fanyue","family":"Wei","sequence":"additional","affiliation":[{"name":"National University of Singapore,Singapore"}]},{"given":"Lixin","family":"Duan","sequence":"additional","affiliation":[{"name":"University of Electronic Science and Technology of China,China"}]},{"given":"Angela","family":"Yao","sequence":"additional","affiliation":[{"name":"National University of Singapore,Singapore"}]},{"given":"Wen","family":"Li","sequence":"additional","affiliation":[{"name":"University of Electronic Science and Technology of China,China"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.00971"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.618"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01270-0_28"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58452-8_13"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.502"},{"key":"ref6","article-title":"Finding moments in video collections using natural language","author":"Escorcia","year":"2019","journal-title":"arXiv preprint arXiv"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00630"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.563"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/WACV.2019.00032"},{"key":"ref10","article-title":"Imagenet-trained cnns are biased towards texture; increasing shape bias improves accuracy and robustness","author":"Geirhos","year":"2019","journal-title":"ICLR"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1007\/s11263-023-01916-5"},{"key":"ref12","article-title":"Understanding the difficulty of training deep feedforward neural networks","author":"Glorot","year":"2010","journal-title":"AISTATS"},{"key":"ref13","article-title":"Model patching: Closing the subgroup performance gap with data augmentation","author":"Goel","year":"2021","journal-title":"ICLR"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV51070.2023.01273"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/WACV61041.2025.00832"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.01280"},{"key":"ref17","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-72627-9_13"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58589-1_27"},{"key":"ref19","first-page":"11846","article-title":"Detecting moments and highlights in videos via natural language queries","author":"Lei","year":"2021","journal-title":"NeurIPS"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2880"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/TMM.2025.3543033"},{"key":"ref22","first-page":"2794","article-title":"Univtg: Towards unified videolanguage temporal grounding","author":"Lin","year":"2023","journal-title":"ICCV"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1145\/3209978.3210003"},{"key":"ref24","doi-asserted-by":"publisher","DOI":"10.1145\/3240508.3240549"},{"key":"ref25","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.00305"},{"key":"ref26","article-title":"Stat: Towards generalizable temporal action localization","author":"Liu","year":"2024","journal-title":"arXiv preprint arXiv"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52734.2025.00814"},{"key":"ref28","article-title":"Decoupled weight decay regularization","author":"Loshchilov","year":"2019","journal-title":"ICLR"},{"key":"ref29","article-title":"Correlation-guided query-dependency calibration for video temporal grounding","author":"Moon","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.02205"},{"key":"ref31","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.01791"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00279"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.5244\/C.34.84"},{"key":"ref34","article-title":"Pytorch: An imperative style, high-performance deep learning library","author":"Paszke","year":"2019","journal-title":"NeurIPS"},{"issue":"198","key":"ref35","first-page":"1","article-title":"Codalab competitions: An open source platform to organize scientific challenges","volume":"24","author":"Pavao","year":"2023","journal-title":"JMLR"},{"key":"ref36","first-page":"8748","article-title":"Learning transferable visual models from natural language supervision","author":"Radford","year":"2021","journal-title":"ICML"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00207"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00075"},{"key":"ref39","article-title":"Distributionally robust neural networks for group shifts: On the importance of regularization for worstcase generalization","author":"Sagawa","year":"2019","journal-title":"arXiv preprint arXiv"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-46448-0_31"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1162\/tacl_a_00335"},{"key":"ref42","first-page":"3745","article-title":"On the effect of key factors in spurious correlation: A theoretical perspective","author":"Wang","year":"2024","journal-title":"AISTATS"},{"key":"ref43","first-page":"3745","article-title":"On the effect of key factors in spurious correlation: A theoretical perspective","volume-title":"International Conference on Artificial Intelligence and Statistics","author":"Wang"},{"key":"ref44","first-page":"12844","article-title":"Next-qa: Next phase of question-answering to explaining temporal and causal video understanding","author":"Xiao","year":"2021","journal-title":"CVPR"},{"key":"ref45","first-page":"13454","article-title":"Unsupervised temporal sentence grounding with cross-modal mutual information maximization","author":"Xu","year":"2021","journal-title":"CVPR"},{"key":"ref46","first-page":"13744","article-title":"Zero-shot video moment retrieval via query-guided video generation","author":"Zang","year":"2025","journal-title":"CVPR"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01030"},{"key":"ref48","first-page":"1123","article-title":"Span-based joint entity and relation extraction with attention-based span-specific representations","author":"Zhang","year":"2023","journal-title":"EMNLP"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i07.6984"},{"key":"ref50","first-page":"4091","article-title":"Regularized two-branch network for weakly-supervised sentence grounding in videos","author":"Zhang","year":"2020","journal-title":"ACM MM"},{"key":"ref51","first-page":"10421","article-title":"Contextualized video grounding with language-guided spatial attention","author":"Zhai","year":"2023","journal-title":"ICCV"},{"key":"ref52","first-page":"5833","article-title":"End-to-end temporal action detection with transformer","author":"Zheng","year":"2021","journal-title":"CVPR"},{"key":"ref53","first-page":"13812","article-title":"Diverse and consistent video grounding with query-guided distillation","author":"Zhao","year":"2025","journal-title":"CVPR"},{"key":"ref54","first-page":"22340","article-title":"Video grounding with spatio-temporal causal reasoning","author":"Zhou","year":"2024","journal-title":"CVPR"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/tpami.2020.3038993"},{"key":"ref56","article-title":"Minigpt-4: Enhancing vision-language understanding with advanced large language models","author":"Zhu","year":"2023","journal-title":"arXiv preprint arXiv"},{"key":"ref57","first-page":"13402","article-title":"Temporal action localization with pyramid avoidance convolution","author":"Zhao","year":"2021","journal-title":"ICCV"},{"key":"ref58","first-page":"1","article-title":"Video grounding with transformer-based span prediction","author":"Zhang","year":"2025","journal-title":"IJCV"},{"key":"ref59","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.317"}],"event":{"name":"2025 IEEE\/CVF International Conference on Computer Vision (ICCV)","location":"Honolulu, HI, USA","start":{"date-parts":[[2025,10,19]]},"end":{"date-parts":[[2025,10,25]]}},"container-title":["2025 IEEE\/CVF International Conference on Computer Vision (ICCV)"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/11443115\/11443287\/11446232.pdf?arnumber=11446232","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T05:21:36Z","timestamp":1777612896000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11446232\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":59,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.01950","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}