arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2603.00823cs.CLcs.AI
收藏

大语言模型多轮交互下LLM去学习鲁棒性综合评估

A Comprehensive Evaluation of LLM Unlearning Robustness under Multi-Turn Interaction

Ruihao Pan, Suhang Wang

AI总结:

本文研究多轮交互环境下LLM去学习的稳定性,发现静态评估可能高估实际效果,强调需确保交互场景下的稳定遗忘。

摘要:

机器去学习旨在不重新训练的情况下移除特定训练数据对预训练模型的影响,这对大语言模型(LLMs)日益重要,因安全、隐私和法律问题。尽管先前工作主要评估静态、单轮设置下的去学习,但现实交互中的遗忘鲁棒性仍待探索。本文通过研究两种常见交互模式:自我纠正和对话条件查询,探讨去学习在交互环境中的稳定性。发现静态评估中遗忘的知识可通过交互恢复。虽然更强的去学习提升表面鲁棒性,但往往导致行为僵化而非真实知识擦除。本文发现静态评估可能高估实际效果,强调需确保交互场景下的稳定遗忘。

英文摘要:

Machine unlearning aims to remove the influence of specific training data from pre-trained models without retraining from scratch, and is increasingly important for large language models (LLMs) due to safety, privacy, and legal concerns. Although prior work primarily evaluates unlearning in static, single-turn settings, forgetting robustness under realistic interactive use remains underexplored. In this paper, we study whether unlearning remains stable in interactive environments by examining two common interaction patterns: self-correction and dialogue-conditioned querying. We find that knowledge appearing forgotten in static evaluation can often be recovered through interaction. Although stronger unlearning improves apparent robustness, it often results in behavioral rigidity rather than genuine knowledge erasure. Our findings suggest that static evaluation may overestimate real-world effectiveness and highlight the need for ensuring stable forgetting under interactive settings.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑