{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T03:02:17Z","timestamp":1773802937099,"version":"3.50.1"},"reference-count":0,"publisher":"Association for the Advancement of Artificial Intelligence (AAAI)","issue":"24","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["AAAI"],"abstract":"<jats:p>Personalized Federated Learning (PFL) customizes models for each client to mitigate challenges from non-IID data, wherein a dominant strategy is model decoupling that partitions models into shared and personalized parts based on architectural priors (e.g., backbone vs. head). However, we reveal a critical flaw in this strategy: it induces \"intrinsic drift,\" a performance degradation often more severe than the well-known client drift, which limits final accuracy. We trace this drift to a steep cliff of high loss emerging from the naive stitching of shared and personalized parts. To address this, we shift from architectural partitioning to a parameter behavior-driven paradigm. We introduce PPFL, an approach that employs a novel soft-fusion strategy guided by parameter-wise behavioral perception. PPFL dynamically infers each parameter's functional role\u2014whether it behaves more like a 'personalist' or a 'generalist' in the current context\u2014by synthesizing its multifaceted behavior observed during local training. Extensive experiments on image, text, and multimodal classification benchmarks show that PPFL outperforms eight state-of-the-art baselines by up to 5.3%. Moreover, it can function as a plug-in module, boosting the accuracy of vanilla FedAvg with a 16.82% absolute gain.<\/jats:p>","DOI":"10.1609\/aaai.v40i24.39073","type":"journal-article","created":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T01:11:20Z","timestamp":1773796280000},"page":"19898-19906","source":"Crossref","is-referenced-by-count":0,"title":["PPFL: A Parameter Behavior-Driven Plug-in Personalization Engine for Federated Learning"],"prefix":"10.1609","volume":"40","author":[{"given":"Qianyue","family":"Cao","sequence":"first","affiliation":[]},{"given":"Zongwei","family":"Zhu","sequence":"additional","affiliation":[]},{"given":"Zirui","family":"Lian","sequence":"additional","affiliation":[]},{"given":"Rui","family":"Zhang","sequence":"additional","affiliation":[]},{"given":"Boyu","family":"Li","sequence":"additional","affiliation":[]},{"given":"Yi","family":"Xiong","sequence":"additional","affiliation":[]},{"given":"Xuehai","family":"Zhou","sequence":"additional","affiliation":[]}],"member":"9382","published-online":{"date-parts":[[2026,3,14]]},"container-title":["Proceedings of the AAAI Conference on Artificial Intelligence"],"original-title":[],"link":[{"URL":"https:\/\/ojs.aaai.org\/index.php\/AAAI\/article\/download\/39073\/43035","content-type":"application\/pdf","content-version":"vor","intended-application":"text-mining"},{"URL":"https:\/\/ojs.aaai.org\/index.php\/AAAI\/article\/download\/39073\/43035","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2026,3,18]],"date-time":"2026-03-18T01:11:21Z","timestamp":1773796281000},"score":1,"resource":{"primary":{"URL":"https:\/\/ojs.aaai.org\/index.php\/AAAI\/article\/view\/39073"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2026,3,14]]},"references-count":0,"journal-issue":{"issue":"24","published-online":{"date-parts":[[2026,3,17]]}},"URL":"https:\/\/doi.org\/10.1609\/aaai.v40i24.39073","relation":{},"ISSN":["2374-3468","2159-5399"],"issn-type":[{"value":"2374-3468","type":"electronic"},{"value":"2159-5399","type":"print"}],"subject":[],"published":{"date-parts":[[2026,3,14]]}}}