Unilogit: Robust Machine Unlearning for LLMs Using Uniform-Target Self-Distillation
It addresses a critical task for compliance with data privacy regulations like GDPR, though it appears incremental as it builds on prior unlearning methods.
This paper tackles the problem of selectively forgetting specific information in Large Language Models while maintaining overall utility, introducing Unilogit, a self-distillation method that outperforms state-of-the-art methods like NPO and UnDIAL on public benchmarks and an in-house e-commerce dataset.
This paper introduces Unilogit, a novel self-distillation method for machine unlearning in Large Language Models. Unilogit addresses the challenge of selectively forgetting specific information while maintaining overall model utility, a critical task in compliance with data privacy regulations like GDPR. Unlike prior methods that rely on static hyperparameters or starting model outputs, Unilogit dynamically adjusts target logits to achieve a uniform probability for the target token, leveraging the current model's outputs for more accurate self-distillation targets. This approach not only eliminates the need for additional hyperparameters but also enhances the model's ability to approximate the golden targets. Extensive experiments on public benchmarks and an in-house e-commerce dataset demonstrate Unilogit's superior performance in balancing forget and retain objectives, outperforming state-of-the-art methods such as NPO and UnDIAL. Our analysis further reveals Unilogit's robustness across various scenarios, highlighting its practical applicability and effectiveness in achieving efficacious machine unlearning.