arXiv / Factual Memory / Token-level

Mem-α: Learning Memory Construction via Reinforcement Learning

作者：Yu Wang, Ryuichi Takanobu, Zhiqi Liang, Yuzhen Mao, Yuanzhe Hu, Julian McAuley, Xiaojian Wu · 来源：arXiv · 发布日期：2025/09

事实记忆词元级记忆
新标签打开 AlphaXiv
Mem-α: Learning Memory Construction via Reinforcement Learning figure
AlphaXiv 中文概览（可滚动查看）