#artificial intelligence
May 2026
EvoMemBench: Benchmarking Agent Memory from a Self-Evolving Perspective
This paper studies agent memory from a self-evolving perspective and introduces EvoMemBench, a unified benchmark organized along two axes: memory scope (in-episode vs. cross-episode) and memory content (knowledge-oriented vs. execution-oriented).
Yu-Yao Wang, Zhongjian Zhang, Mo-Han Chi et al.
· arXiv.org · 21 citations
· ⚡1