TL;DR
Oblivionis is a lightweight framework that enables federated large language models to selectively forget private data, improving privacy compliance without sacrificing model performance.
Contribution
It introduces a dual optimization approach unifying federated learning and unlearning, with comprehensive algorithms for effective data removal in federated LLMs.
Findings
Outperforms local training in forgetting and utility balance
Provides a robust pipeline for federated LLM unlearning
Offers comparative analysis of multiple algorithms
Abstract
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…
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
