1
ToolFailBench: Diagnosing Tool-Use Failures in LLM Agents
新基准ToolFailBench精准诊断LLM agent工具调用失败原因，揭示聚合分数下的隐藏缺陷。
arXiv:2607.04686v1 Announce Type: cross Abstract: Tool calling is central to modern language model agents, but aggregate benchmark scores often hide w…