MadeAgents/HammerBench
收藏官方服务：
资源简介：
HammerBench是一个针对真实世界移动设备场景下的细粒度函数调用评估的基准数据集。它包含单轮和多轮对话，能够评估大型语言模型在不同情况下（如不完整指令、多变的问答轨迹、意图/参数转换、外部个人信息引用）的性能。
HammerBench is a benchmark for fine-grained function-calling evaluation in real mobile device scenarios. It includes single-turn and multi-turn dialogues and is designed to evaluate the performance of large language models under various conditions such as imperfect instructions, diverse question-answer trajectories, intent/argument shifts, and references to external individual information.
提供机构：
MadeAgents
