{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,5,4]],"date-time":"2026-05-04T10:15:55Z","timestamp":1777889755565,"version":"3.51.4"},"reference-count":74,"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"}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":[],"published-print":{"date-parts":[[2025,10,19]]},"DOI":"10.1109\/iccv51701.2025.00081","type":"proceedings-article","created":{"date-parts":[[2026,4,29]],"date-time":"2026-04-29T19:45:49Z","timestamp":1777491949000},"page":"785-795","source":"Crossref","is-referenced-by-count":0,"title":["SAMO: A Lightweight Sharpness-Aware Approach for Multi-Task Optimization with Joint Global-Local Perturbation"],"prefix":"10.1109","author":[{"given":"Hao","family":"Ban","sequence":"first","affiliation":[{"name":"University at Buffalo"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Gokul Ram","family":"Subramani","sequence":"additional","affiliation":[{"name":"University at Buffalo"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Kaiyi","family":"Ji","sequence":"additional","affiliation":[{"name":"University at Buffalo"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","first-page":"10","article-title":"Sharp-maml: Sharpness-aware model-agnostic meta learning","volume-title":"International Conference on Machine Learning","author":"Abbas"},{"key":"ref2","first-page":"639","article-title":"Towards understanding sharpness-aware minimization","volume-title":"International Conference on Machine Learning","author":"Andriushchenko"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2037"},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2016.2644615"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2022.acl-long.508"},{"key":"ref6","first-page":"2715","article-title":"Fair resource allocation in multi-task learning","volume-title":"International Conference on Machine Learning","volume":"1","author":"Ban"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.5244\/c.34.93"},{"key":"ref8","first-page":"22405","article-title":"Swad: Domain generalization by seeking flat minima","volume":"34","author":"Cha","year":"2021","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/3663363"},{"key":"ref10","article-title":"Gradient-based multi-objective deep learning: Algorithms, theories, applications, and beyond","author":"Chen","year":"2025","journal-title":"arXiv preprint"},{"key":"ref11","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"},{"key":"ref12","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":"ref13","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01138"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.52202\/075280-3165"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1016\/j.neucom.2019.11.023"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.350"},{"key":"ref17","article-title":"Multi-task learning with deep neural networks: A survey","author":"Crawshaw","year":"2020","journal-title":"arXiv preprint"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.343"},{"key":"ref19","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2009.5206848"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1016\/j.crma.2012.03.014"},{"key":"ref21","first-page":"1019","article-title":"Sharp minima can generalize for deep nets","volume-title":"International Conference on Machine Learning","author":"Dinh"},{"key":"ref22","article-title":"Efficient sharpness-aware minimization for improved training of neural networks","volume-title":"International Conference on Learning Representations","author":"Du"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.52202\/068431-2062"},{"key":"ref24","article-title":"Mitigating gradient bias in multi-objective learning: A provably convergent approach","volume-title":"The Eleventh International Conference on Learning Representations","volume":"2","author":"Devaka Fernando"},{"key":"ref25","article-title":"Fast graph representation learning with pytorch geometric","author":"Fey","year":"2019","journal-title":"arXiv preprint"},{"key":"ref26","article-title":"Sharpness-aware minimization for efficiently improving generalization","volume-title":"International Conference on Learning Representations","volume":"1","author":"Foret"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00332"},{"key":"ref28","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR42600.2020.01156"},{"key":"ref29","article-title":"Variance-reduced zeroth-order methods for fine-tuning language models","volume-title":"Forty-first International Conference on Machine Learning.","author":"Gautam"},{"key":"ref30","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2016.90"},{"key":"ref31","article-title":"Simplifying neural nets by discovering flat minima","volume":"7","author":"Hochreiter","year":"1994","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref32","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.1.1"},{"key":"ref33","article-title":"Lora: Lowrank adaptation of large language models","volume-title":"International Conference on Learning Representations","author":"Hu"},{"key":"ref34","article-title":"The break-even point on optimization trajectories of deep neural networks","volume-title":"International Conference on Learning Representations","author":"Jastrzebski"},{"key":"ref35","article-title":"Rotograd: Gradient homogenization in multitask learning","volume-title":"International Conference on Learning Representations","author":"Javaloy"},{"key":"ref36","article-title":"Fantastic generalization measures and where to find them","volume-title":"International Conference on Learning Representations","author":"Jiang"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00781"},{"key":"ref38","article-title":"On largebatch training for deep learning: Generalization gap and sharp minima","volume-title":"International Conference on Learning Representations","author":"Shirish Keskar"},{"key":"ref39","article-title":"Visualizing the loss landscape of neural nets","volume":"31","author":"Li","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"issue":"209","key":"ref40","first-page":"1","article-title":"LibMTL: A Python library for multi-task learning","volume":"24","author":"Lin","year":"2023","journal-title":"Journal of Machine Learning Research"},{"key":"ref41","article-title":"Reasonable effectiveness of random weighting: A litmus test for multi-task learning","volume":"2","author":"Lin","journal-title":"Transactions on Machine Learning Research."},{"key":"ref42","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":"ref43","doi-asserted-by":"publisher","DOI":"10.52202\/075280-2500"},{"key":"ref44","article-title":"Towards impartial multi-task learning","volume-title":"International Conference on Learning Representations.","author":"Liu"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2019.00197"},{"key":"ref46","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01204"},{"key":"ref47","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.425"},{"key":"ref48","first-page":"16428","article-title":"Multitask learning as a bargaining game","volume-title":"International Conference on Machine Learning","author":"Navon"},{"key":"ref49","article-title":"Improving multi-task learning via seeking task-based flat regions","author":"Phan","year":"2022","journal-title":"arXiv preprint"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1038\/sdata.2014.22"},{"key":"ref51","article-title":"An overview of multi-task learning in deep neural networks","author":"Ruder","year":"2017","journal-title":"arXiv preprint"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v33i01.33014822"},{"key":"ref53","article-title":"Multi-task learning as multi-objective optimization","volume":"31","author":"Sener","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref54","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.01923"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-642-33715-4_54"},{"key":"ref56","doi-asserted-by":"publisher","DOI":"10.1109\/9.119632"},{"key":"ref57","article-title":"Sharpness-aware minimization enhances feature quality via balanced learning","volume-title":"The Twelfth International Conference on Learning Representations","author":"Mitchell Springer"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/ICASSP48485.2024.10445941"},{"issue":"11","key":"ref59","article-title":"Visualizing data using t -sne","volume":"9","author":"Van der Maaten","year":"2008","journal-title":"Journal of Machine Learning Research"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3054719"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.572"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52729.2023.00367"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1109\/TASLP.2024.3434425"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2025.3581454"},{"key":"ref65","article-title":"How sharpnessaware minimization minimizes sharpness?","volume-title":"The Eleventh International Conference on Learning Representations","author":"Wen"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.52202\/075280-0200"},{"key":"ref67","article-title":"Scalable bilevel loss balancing for multi-task learning","author":"Xiao","year":"2025","journal-title":"arXiv preprint"},{"key":"ref68","article-title":"Hessian-based analysis of large batch training and robustness to adversaries","volume":"31","author":"Yao","year":"2018","journal-title":"Advances in Neural Information Processing Systems"},{"key":"ref69","article-title":"Unleashing the power of multitask learning: A comprehensive survey spanning traditional, deep, and pretrained foundation model eras","author":"Yu","year":"2024","journal-title":"arXiv preprint"},{"key":"ref70","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":"ref71","first-page":"arXiv-2405","article-title":"On the convergence of multi-objective optimization under generalized smoothness","author":"Zhang","year":"2024","journal-title":"arXiv e-prints"},{"key":"ref72","doi-asserted-by":"publisher","DOI":"10.1109\/TKDE.2021.3070203"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.18653\/v1\/2023.eacl-main.66"},{"key":"ref74","article-title":"Surrogate gap minimization improves sharpnessaware training","volume-title":"International Conference on Learning Representations","author":"Zhuang"}],"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\/11445554.pdf?arnumber=11445554","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,5,1]],"date-time":"2026-05-01T05:18:10Z","timestamp":1777612690000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11445554\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10,19]]},"references-count":74,"URL":"https:\/\/doi.org\/10.1109\/iccv51701.2025.00081","relation":{},"subject":[],"published":{"date-parts":[[2025,10,19]]}}}