Oblivionis: A Lightweight Learning and Unlearning Framework for Federated Large Language Models
Proceedings of the AAAI Conference on Artificial Intelligence · 14.03.2026 · 40 (33) · 28185-28193
DOI: 10.1609/aaai.v40i33.40045 ↗Kurz gefasst
Fachartikel in der Zeitschrift Proceedings of the AAAI Conference on Artificial Intelligence, online erschienen am 14.03.2026. Verfasst von Fuyao Zhang und 9 weiteren Autor:innen. Die Arbeit wurde bisher 1-mal zitiert (Crossref).
Abstract (Originalsprache)
Large Language Models (LLMs) increasingly leverage Federated Learning (FL) to utilize private, task-specific datasets for fine-tuning while preserving data privacy. However, while federated LLM frameworks effectively enable collaborative training without raw data sharing, they critically lack built-in mechanisms for regulatory compliance like GDPR’s right to be forgotten. Integrating private data heightens concerns over data quality and long-term governance, yet existing distributed training frameworks offer no principled way to selectively remove specific client contributions post-training. Due to distributed data silos, stringent privacy constraints, and the intricacies of interdependent model aggregation, federated LLM unlearning is significantly more complex than centralized LLM unlearning. To address this gap, we introduce Oblivionis, a lightweight learning and unlearning framework that enables clients to selectively remove specific private data during federated LLM training, enhancing trustworthiness and regulatory compliance. By unifying FL and unlearning as a dual optimization objective, we incorporate 6 FL and 5 unlearning algorithms for comprehensive evaluation and comparative analysis, establishing a robust pipeline for federated LLM unlearning. Extensive experiments demonstrate that Oblivionis outperforms local training, achieving a robust balance between forgetting efficacy and model utility, with cross-algorithm comparisons providing clear directions for future LLM development.
Zitieren
APA
Zhang, F., Yan, X., Wu, T., Li, W., Chen, T., Cao, Y., Yan, R., Huang, L., Lim, W. Y. B., & Yang, Q. (2026). Oblivionis: A Lightweight Learning and Unlearning Framework for Federated Large Language Models. Proceedings of the AAAI Conference on Artificial Intelligence, 40(33), 28185-28193. https://doi.org/10.1609/aaai.v40i33.40045
Vancouver
Zhang F, Yan X, Wu T, Li W, Chen T, Cao Y, et al. Oblivionis: A Lightweight Learning and Unlearning Framework for Federated Large Language Models. Proceedings of the AAAI Conference on Artificial Intelligence. 2026;40(33):28185-28193. doi:10.1609/aaai.v40i33.40045
BibTeX
@article{Zhang2026,
author = {Zhang, Fuyao and Yan, Xinyu and Wu, Tiantong and Li, Wenjie and Chen, Tianxiang and Cao, Yang and Yan, Ran and Huang, Longtao and Lim, Wei Yang Bryan and Yang, Qiang},
title = {Oblivionis: A Lightweight Learning and Unlearning Framework for Federated Large Language Models},
journal = {Proceedings of the AAAI Conference on Artificial Intelligence},
year = {2026},
volume = {40},
number = {33},
pages = {28185--28193},
doi = {10.1609/aaai.v40i33.40045}
}
EUVIMED