{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T10:09:33Z","timestamp":1777889373880,"version":"3.51.4"},"reference-count":72,"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\/501100003626","name":"Defense Acquisition Program Administration","doi-asserted-by":"publisher","award":["915102201"],"award-info":[{"award-number":["915102201"]}],"id":[{"id":"10.13039\/501100003626","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\/iccv51701.2025.00277","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"2887-2897","source":"Crossref","is-referenced-by-count":0,"title":["Resolving Token-Space Gradient Conflicts: Token Space Manipulation for Transformer-Based Multi-Task Learning"],"prefix":"10.1109","author":[{"given":"Wooseong","family":"Jeong","sequence":"first","affiliation":[{"name":"KAIST"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kuk-Jin","family":"Yoon","sequence":"additional","affiliation":[{"name":"KAIST"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","doi-asserted-by":"publisher","DOI":"10.1109\/cvpr52688.2022.01172"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00147"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.5244\/C.34.93"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.5244\/C.34.93"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1109\/CEC45853.2021.9504721"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1023\/A:1007379606734"},{"key":"ref7","first-page":"794","article-title":"Gradnorm: Gradient normalization for adaptive loss balancing in deep multitask networks","volume-title":"International conference on machine learning","author":"Chen","year":"2018"},{"key":"ref8","first-page":"2039","article-title":"Just pick a sign: Optimizing deep multitask models with gradient sign dropout","volume":"33","author":"Chen","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01138"},{"key":"ref10","first-page":"1","article-title":"Multi-task learning with deep neural networks: A survey","author":"Crawshaw","year":"2020","journal-title":"arXiv"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.343"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref13","doi-asserted-by":"publisher","DOI":"10.1016\/j.crma.2012.03.014"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2010.11929"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.01267"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.304"},{"key":"ref17","first-page":"1","article-title":"The pascal visual object classes challenge 2012 (voc2012) development kit","volume":"2007","author":"Everingham","year":"2012","journal-title":"Pattern Anal. Stat. Model. Comput. Learn., Tech. Rep"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.52202\/068431-2062"},{"key":"ref19","first-page":"12","article-title":"Pathnet: Evolution channels gradient descent in super neural networks","author":"Fernando","year":"2017","journal-title":"arXiv"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00332"},{"key":"ref21","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01156"},{"key":"ref22","article-title":"Recon: Reducing conflicting gradients from the root for multi-task learning","volume-title":"The Eleventh International Conference on Learning Representations","author":"Guangyuan","year":"2022"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01270-0_17"},{"key":"ref24","first-page":"3854","article-title":"Learning to branch for multi-task learning","volume-title":"International conference on machine learning","author":"Guo","year":"2020"},{"key":"ref25","first-page":"3854","article-title":"Learning to branch for multi-task learning","volume-title":"International Conference on Machine Learning","author":"Guo","year":"2020"},{"issue":"2","key":"ref26","first-page":"3","article-title":"Lora: Low-rank adaptation of large language models","volume":"1","author":"Hu","year":"2022","journal-title":"ICLR"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.02662"},{"key":"ref28","first-page":"3","article-title":"Rotograd: Gradient homogenization in multitask learning","author":"Javaloy","year":"2021","journal-title":"arXiv"},{"key":"ref29","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52733.2024.00042"},{"key":"ref30","first-page":"3","article-title":"Selective task group updates for multi-task optimization","author":"Jeong","year":"2025","journal-title":"arXiv"},{"key":"ref31","article-title":"Forkmerge: Mitigating negative transfer in auxiliary-task learning","volume":"36","author":"Jiang","year":"2024","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1098\/rsta.2015.0202"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00781"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1145\/3205455.3205489"},{"key":"ref35","first-page":"18878","article-title":"Conflict-averse gradient descent for multi-task learning","volume":"34","author":"Liu","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2500"},{"key":"ref37","article-title":"Towards impartial multi-task learning","author":"Liu","year":"2021","journal-title":"iclr"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00197"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00986"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.126"},{"key":"ref41","doi-asserted-by":"publisher","DOI":"10.1145\/3219819.3220007"},{"key":"ref42","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00195"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2014.119"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.52202\/068431-0695"},{"key":"ref45","article-title":"Multitask learning as a bargaining game","author":"Navon","year":"2022","journal-title":"arXiv"},{"key":"ref46","first-page":"512","article-title":"What is being transferred in transfer learning?","volume":"33","author":"Neyshabur","year":"2020","journal-title":"Advances in neural information processing systems"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/P19-1185"},{"key":"ref48","first-page":"8583","article-title":"Scaling vision with sparse mixture of experts","volume":"34","author":"Riquelme","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref49","article-title":"Multi-task learning as multi-objective optimization","volume":"31","author":"Sener","year":"2018","journal-title":"Advances in neural information processing systems"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01923"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-33715-4_54"},{"key":"ref52","first-page":"2","article-title":"One-shot neural architecture search for deep multi-task learning in computer vision","author":"Silvestri","year":"2020","journal-title":"One-shot neural architecture search for deep multi-task learning in computer vision"},{"key":"ref53","doi-asserted-by":"publisher","DOI":"10.48550\/ARXIV.1409.1556"},{"key":"ref54","first-page":"12","article-title":"Gradient adversarial training of neural networks","author":"Sinha","year":"2018","journal-title":"Gradient adversarial training of neural networks."},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00818"},{"key":"ref56","first-page":"12","article-title":"Branched multi-task networks: deciding what layers to share","author":"Vandenhende","year":"2019","journal-title":"arXiv"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58548-8_31"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00061"},{"key":"ref59","first-page":"2","article-title":"Crossformer: A versatile vision transformer hinging on cross-scale attention","author":"Wang","year":"2021","journal-title":"arXiv"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.48550\/arXiv.2105.15203"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00077"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19812-0_18"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/TCSVT.2023.3292995"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v37i3.25411"},{"key":"ref65","first-page":"2","article-title":"Focal self-attention for local-global interactions in vision transformers","author":"Yang","year":"2021","journal-title":"arXiv"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-031-19812-0_30"},{"key":"ref67","article-title":"Taskprompter: Spatial-channel multi-task prompting for dense scene understanding","volume-title":"The Eleventh International Conference on Learning Representations","author":"Ye","year":"2022"},{"key":"ref68","first-page":"5824","article-title":"Gradient surgery for multi-task learning","volume":"33","author":"Yu","year":"2020","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00391"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.emnlp-main.278"},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-10599-4_7"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00423"}],"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\/11444843.pdf?arnumber=11444843","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T05:10:52Z","timestamp":1777612252000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11444843\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":72,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.00277","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}