Paper ID: 2404.12045

RAM: Towards an Ever-Improving Memory System by Learning from Communications

Jiaqi Li, Xiaobo Wang, Wentao Ding, Zihao Wang, Yipeng Kang, Zixia Jia, Zilong Zheng

We introduce an innovative RAG-based framework with an ever-improving memory. Inspired by humans'pedagogical process, RAM utilizes recursively reasoning-based retrieval and experience reflections to continually update the memory and learn from users' communicative feedback, namely communicative learning. Extensive experiments with both simulated and real users demonstrate significant improvements over traditional RAG and self-knowledge methods, particularly excelling in handling false premise and multi-hop questions. Furthermore, RAM exhibits promising adaptability to various feedback and retrieval methods, showcasing its potential for advancing AI capabilities in dynamic knowledge acquisition and lifelong learning.

Submitted: Apr 18, 2024