1
PM-Bench: Evaluating Prospective Memory in LLM Agents
首个评估LLM智能体“前瞻性记忆”的基准，已被COLM 2026接收，为Agent记忆研究提供新测试标准。
arXiv:2607.12385v1 Announce Type: new Abstract: A significant challenge in agentic AI is prospective memory: the ability to execute an intention at a …