作者位次 #6/9ACL 2026Poster· Natural Language Processing
FOREVER : Forgetting Curve-Inspired Memory Replay for Language Model Continual Learning
Yujie Feng · Hao Wang · Jian Li · Xu Chu · Zhaolu Kang · Yiran Liu · Yasha Wang · Philip S. Yu · Xiao-Ming Wu
摘要
Continual learning (CL) for large language models (LLMs) aims to enable sequential knowledge acquisition without catastrophic forgetting. Memory replay methods are widely used for their practicality and effectiveness, but most rely on fixed, step-based heuristics that often misalign with the mode...