reinforcement-policy

Tag

Cards List
#reinforcement-policy

CAMR: Compatibility-Aware Memory Replay for Continual Learning in Large Language Models

arXiv cs.CL ↗ · 5d ago Cached

论文提出 CAMR(兼容性感知记忆回放)框架,通过兼容性检查与学习到的记忆门控联合决定当前样本的监督与历史样本回放策略,在 Qwen3-8B-Base 上实现了持续事实更新与跨任务学习的更优稳定性和保持能力。

0 favorites 0 likes
← Back to home

Submit Feedback