{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,10,9]],"date-time":"2025-10-09T01:04:34Z","timestamp":1759971874248,"version":"build-2065373602"},"reference-count":46,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"10","license":[{"start":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T00:00:00Z","timestamp":1759276800000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T00:00:00Z","timestamp":1759276800000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,10,1]],"date-time":"2025-10-01T00:00:00Z","timestamp":1759276800000},"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":["62403484","62402425","62306324"],"award-info":[{"award-number":["62403484","62402425","62306324"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100017940","name":"Zhejiang Provincial Priority-Funded Postdoctoral Research Project","doi-asserted-by":"publisher","award":["ZJ2024001"],"award-info":[{"award-number":["ZJ2024001"]}],"id":[{"id":"10.13039\/100017940","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Neural Netw. Learning Syst."],"published-print":{"date-parts":[[2025,10]]},"DOI":"10.1109\/tnnls.2025.3585063","type":"journal-article","created":{"date-parts":[[2025,7,17]],"date-time":"2025-07-17T17:59:53Z","timestamp":1752775193000},"page":"19054-19067","source":"Crossref","is-referenced-by-count":0,"title":["The Risk of Federated Learning to Skew Fine-Tuning Features and Underperform Robustness"],"prefix":"10.1109","volume":"36","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-9560-2869","authenticated-orcid":false,"given":"Mengyao","family":"Du","sequence":"first","affiliation":[{"name":"School of Systems Engineering, National University of Defense Technology, Changsha, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-1839-3079","authenticated-orcid":false,"given":"Miao","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Systems Engineering, National University of Defense Technology, Changsha, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-2311-4943","authenticated-orcid":false,"given":"Yuwen","family":"Pu","sequence":"additional","affiliation":[{"name":"School of Big Data and Software Engineering, Chongqing University, Chongqing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-6085-7300","authenticated-orcid":false,"given":"Qingming","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer Science, Zhejiang University, Hangzhou, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-4268-372X","authenticated-orcid":false,"given":"Shouling","family":"Ji","sequence":"additional","affiliation":[{"name":"School of Computer Science, Zhejiang University, Hangzhou, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-1207-8660","authenticated-orcid":false,"given":"Quanjun","family":"Yin","sequence":"additional","affiliation":[{"name":"School of Systems Engineering, National University of Defense Technology, Changsha, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"key":"ref1","article-title":"BERT: Pre-training of deep bidirectional transformers for language understanding","author":"Devlin","year":"2018","journal-title":"arXiv:1810.04805"},{"issue":"1","key":"ref2","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":"ref3","article-title":"LLaMA: Open and efficient foundation language models","author":"Touvron","year":"2023","journal-title":"arXiv:2302.13971"},{"key":"ref4","article-title":"Understanding and mitigating the label noise in pre-training on downstream tasks","author":"Chen","year":"2023","journal-title":"arXiv:2309.17002"},{"key":"ref5","article-title":"Revisiting out-of-distribution robustness in NLP: Benchmark, analysis, and LLMs evaluations","author":"Yuan","year":"2023","journal-title":"arXiv:2306.04618"},{"key":"ref6","article-title":"The future of large language model pre-training is federated","author":"Sani","year":"2024","journal-title":"arXiv:2405.10853"},{"doi-asserted-by":"publisher","key":"ref7","DOI":"10.1007\/978-3-030-80599-9_2"},{"key":"ref8","article-title":"Prefix-tuning: Optimizing continuous prompts for generation","author":"Lisa Li","year":"2021","journal-title":"arXiv:2101.00190"},{"key":"ref9","first-page":"2790","article-title":"Parameter-efficient transfer learning for NLP","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Houlsby"},{"key":"ref10","article-title":"LoRA: Low-rank adaptation of large language models","author":"Hu","year":"2021","journal-title":"arXiv:2106.09685"},{"key":"ref11","article-title":"BitFit: Simple parameter-efficient fine-tuning for transformer-based masked language-models","author":"Zaken","year":"2021","journal-title":"arXiv:2106.10199"},{"key":"ref12","first-page":"10","article-title":"Domain generalization via invariant feature representation","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Muandet"},{"doi-asserted-by":"publisher","key":"ref13","DOI":"10.1109\/TPAMI.2022.3195549"},{"doi-asserted-by":"publisher","key":"ref14","DOI":"10.1109\/CVPR52688.2022.00779"},{"doi-asserted-by":"publisher","key":"ref15","DOI":"10.1109\/tnnls.2025.3526235"},{"key":"ref16","first-page":"1877","article-title":"Language models are few-shot learners","volume-title":"Proc. NIPS","author":"Brown"},{"doi-asserted-by":"publisher","key":"ref17","DOI":"10.1109\/CVPR52729.2023.01185"},{"key":"ref18","article-title":"Exploring the impact of instruction data scaling on large language models: An empirical study on real-world use cases","author":"Ji","year":"2023","journal-title":"arXiv:2303.14742"},{"key":"ref19","first-page":"55006","article-title":"LIMA: Less is more for alignment","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"36","author":"Zhou"},{"doi-asserted-by":"publisher","key":"ref20","DOI":"10.1109\/CVPR52688.2022.00780"},{"key":"ref21","first-page":"3495","article-title":"Robust weight signatures: Gaining robustness as easy as patching weights?","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Cai"},{"key":"ref22","first-page":"11905","article-title":"Out-of-distribution generalization of federated learning via implicit invariant relationships","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Guo"},{"key":"ref23","article-title":"Towards understanding and mitigating dimensional collapse in heterogeneous federated learning","author":"Shi","year":"2022","journal-title":"arXiv:2210.00226"},{"key":"ref24","first-page":"5972","article-title":"No fear of heterogeneity: Classifier calibration for federated learning with non-IID data","volume-title":"Proc. NIPS","author":"Luo"},{"doi-asserted-by":"publisher","key":"ref25","DOI":"10.1109\/TNNLS.2024.3439878"},{"key":"ref26","first-page":"1","article-title":"Does learning from decentralized non-IID unlabeled data benefit from self supervision?","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Wang"},{"doi-asserted-by":"publisher","key":"ref27","DOI":"10.1109\/TNNLS.2024.3517658"},{"key":"ref28","first-page":"429","article-title":"Federated optimization in heterogeneous networks","volume-title":"Proc. Mach. Learn. Syst.","volume":"2","author":"Li"},{"doi-asserted-by":"publisher","key":"ref29","DOI":"10.1109\/TNNLS.2024.3498460"},{"doi-asserted-by":"publisher","key":"ref30","DOI":"10.1109\/TNNLS.2024.3438281"},{"key":"ref31","first-page":"6357","article-title":"Ditto: Fair and robust federated learning through personalization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Li"},{"key":"ref32","article-title":"FederatedScope-LLM: A comprehensive package for fine-tuning large language models in federated learning","author":"Kuang","year":"2023","journal-title":"arXiv:2309.00363"},{"key":"ref33","article-title":"SLoRA: Federated parameter efficient fine-tuning of language models","author":"Babakniya","year":"2023","journal-title":"arXiv:2308.06522"},{"key":"ref34","first-page":"24851","article-title":"Investigating why contrastive learning benefits robustness against label noise","volume-title":"Proc. Int. Conf. Mach. Learn. (ICML)","author":"Xue"},{"key":"ref35","first-page":"1081","article-title":"Transferability vs. discriminability: Batch spectral penalization for adversarial domain adaptation","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Chen"},{"doi-asserted-by":"publisher","key":"ref36","DOI":"10.1109\/CVPR52688.2022.00982"},{"doi-asserted-by":"publisher","key":"ref37","DOI":"10.1145\/2507157.2507163"},{"key":"ref38","article-title":"DynaSent: A dynamic benchmark for sentiment analysis","author":"Potts","year":"2020","journal-title":"arXiv:2012.15349"},{"doi-asserted-by":"publisher","key":"ref39","DOI":"10.1007\/978-3-642-35176-1_32"},{"doi-asserted-by":"publisher","key":"ref40","DOI":"10.18653\/v1\/D13-1170"},{"doi-asserted-by":"publisher","key":"ref41","DOI":"10.18653\/v1\/2023.findings-acl.632"},{"doi-asserted-by":"publisher","key":"ref42","DOI":"10.48550\/ARXIV.1907.11692"},{"doi-asserted-by":"publisher","key":"ref43","DOI":"10.18653\/v1\/2022.findings-naacl.13"},{"key":"ref44","first-page":"1950","article-title":"Few-shot parameter-efficient fine-tuning is better and cheaper than in-context learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Liu"},{"key":"ref45","first-page":"1273","article-title":"Communication-efficient learning of deep networks from decentralized data","volume-title":"Proc. 20th Int. Conf. Artif. Intell. Statist.","author":"McMahan"},{"issue":"100","key":"ref46","first-page":"1","article-title":"FedLab: A flexible federated learning framework","volume":"24","author":"Zeng","year":"2023","journal-title":"J. Mach. Learn. Res."}],"container-title":["IEEE Transactions on Neural Networks and Learning Systems"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/5962385\/11195929\/11083121.pdf?arnumber=11083121","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,10,8]],"date-time":"2025-10-08T17:39:09Z","timestamp":1759945149000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/11083121\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,10]]},"references-count":46,"journal-issue":{"issue":"10"},"URL":"https:\/\/doi.org\/10.1109\/tnnls.2025.3585063","relation":{},"ISSN":["2162-237X","2162-2388"],"issn-type":[{"type":"print","value":"2162-237X"},{"type":"electronic","value":"2162-2388"}],"subject":[],"published":{"date-parts":[[2025,10]]}}}