TensorX
返回文献探索

Paper · arXiv 2502.11190

ReLearn: Unlearning via Learning for Large Language Models

Haoming Xu, Ningyuan Zhao, Liming Yang, Sendong Zhao, Shumin Deng, Mengru Wang, Bryan Hooi, Nay Oo, Huajun Chen, Ningyu Zhang

30 upvotesFebruary 16, 2025arXiv 预印本
AI 摘要

ReLearn mitigates issues in current unlearning methods for large language models by proposing a data augmentation and fine-tuning pipeline that effectively targets forgetting while preserving quality and coherence, using novel metrics for evaluation.

reverse optimizationunlearningdata augmentationfine-tuningKnowledge Forgetting RateKnowledge Retention RateLinguistic Scorecoherent text generation

Abstract

Current unlearning methods for large language models usually rely on reverse optimization to reduce target token probabilities. However, this paradigm disrupts the subsequent tokens prediction, degrading model performance and linguistic coherence. Moreover, existing evaluation metrics overemphasize contextual forgetting while inadequately assessing response fluency and relevance. To address these challenges, we propose ReLearn, a data augmentation and fine-tuning pipeline for effective unlearning, along with a comprehensive evaluation framework. This framework introduces Knowledge Forgetting Rate (KFR) and Knowledge Retention Rate (KRR) to measure knowledge-level preservation, and Linguistic Score (LS) to evaluate generation quality. Our experiments show that ReLearn successfully achieves targeted forgetting while preserving high-quality output. Through mechanistic analysis, we further demonstrate how reverse optimization disrupts coherent text generation, while ReLearn preserves this essential capability. Code is available at https://github.com/zjunlp/unlearn.

北京市昌平区探索星信息技术及软件开发工作室

京ICP备2026059466号
ReLearn: Unlearning via Learning for Large Language Models | TensorX