{"ID":2856777,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2510.15967","arxiv_id":"2510.15967","title":"Gains: Fine-grained Federated Domain Adaptation in Open Set","abstract":"Conventional federated learning (FL) assumes a closed world with a fixed total number of clients. In contrast, new clients continuously join the FL process in real-world scenarios, introducing new knowledge. This raises two critical demands: detecting new knowledge, i.e., knowledge discovery, and integrating it into the global model, i.e., knowledge adaptation. Existing research focuses on coarse-grained knowledge discovery, and often sacrifices source domain performance and adaptation efficiency. To this end, we propose a fine-grained federated domain adaptation approach in open set (Gains). Gains splits the model into an encoder and a classifier, empirically revealing features extracted by the encoder are sensitive to domain shifts while classifier parameters are sensitive to class increments. Based on this, we develop fine-grained knowledge discovery and contribution-driven aggregation techniques to identify and incorporate new knowledge. Additionally, an anti-forgetting mechanism is designed to preserve source domain performance, ensuring balanced adaptation. Experimental results on multi-domain datasets across three typical data-shift scenarios demonstrate that Gains significantly outperforms other baselines in performance for both source-domain and target-domain clients. Code is available at: https://github.com/Zhong-Zhengyi/Gains.","short_abstract":"Conventional federated learning (FL) assumes a closed world with a fixed total number of clients. In contrast, new clients continuously join the FL process in real-world scenarios, introducing new knowledge. This raises two critical demands: detecting new knowledge, i.e., knowledge discovery, and integrating it into th...","url_abs":"https://arxiv.org/abs/2510.15967","url_pdf":"https://arxiv.org/pdf/2510.15967v1","authors":"[\"Zhengyi Zhong\",\"Wenzheng Jiang\",\"Weidong Bao\",\"Ji Wang\",\"Cheems Wang\",\"Guanbo Wang\",\"Yongheng Deng\",\"Ju Ren\"]","published":"2025-10-12T13:38:11Z","proceeding":"cs.LG","tasks":"[\"cs.LG\",\"cs.AI\"]","methods":"[]","has_code":false,"code_links":[{"ID":608382,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_id":2856777,"paper_url":"https://arxiv.org/abs/2510.15967","paper_title":"Gains: Fine-grained Federated Domain Adaptation in Open Set","repo_url":"https://github.com/Zhong-Zhengyi/Gains","is_official":false,"mentioned_in_paper":false,"mentioned_in_github":true,"github_stars":0}]}