作者位次 #2/7ACL 2026Poster· Natural Language Processing
Modeling LLM Unlearning as an Asymmetric Two-Task Learning Problem
Zeguan Xiao · Siqing Li · Yong Wang · Xuetao Wei · Jian Yang · Yun Chen · Guanhua Chen
摘要
Machine unlearning for large language models (LLMs) aims to remove targeted knowledge while preserving general capability. In this paper, we recast LLM unlearning as an asymmetric two-task problem: retention is the primary objective and forgetting is an auxiliary. From this perspective, we propos...