EvoMemBench：从自进化视角评测智能体记忆能力

View PDF HTML (experimental)

Abstract:Recent benchmarks for Large Language Model (LLM) agents mainly evaluate reasoning, planning, and execution. However, memory is also essential for agents, as it enables them to store, update, and retrieve information over time. This ability remains under-evaluated, largely because existing benchmarks do not provide a systematic way to assess memory mechanisms. In this paper, we study agent memory from a self-evolving perspective and introduce EvoMemBench, a unified benchmark organized along two axes: memory scope (in-episode vs. cross-episode) and memory content (knowledge-oriented vs. execution-oriented). We compare 15 representative memory methods with strong long-context baselines under a standardized protocol. Results show that current memory systems are still far from a general solution: long-context baselines remain highly competitive, memory helps most when the current context is insufficient or tasks are difficult, and no single memory form works consistently across all settings. Retrieval-based methods remain strong for knowledge-intensive settings, whereas procedural and long-term memory methods are more effective for execution-oriented tasks when their stored experience matches the task structure. We hope EvoMemBench facilitates future research on more effective memory systems for LLM-based agents. Our dataset and code are available at this https URL.

Comments:	Accepted by NeurIPS 2026
Subjects:	Computation and Language (cs.CL); Artificial Intelligence (cs.AI); Machine Learning (cs.LG)
Cite as:	arXiv:2605.18421 [cs.CL]
	(or arXiv:2605.18421v3 [cs.CL] for this version)
	https://doi.org/10.48550/arXiv.2605.18421 arXiv-issued DOI via DataCite

Submission history

From: Yuyao Wang [view email]
[v1] Mon, 18 May 2026 13:54:38 UTC (571 KB)
[v2] Mon, 15 Jun 2026 07:27:08 UTC (566 KB)
[v3] Fri, 30 Oct 2026 08:21:19 UTC (575 KB)