antgroup/HumanSense_Benchmark
收藏官方服务：
资源简介：
HumanSense是一个旨在评估多模态大型语言模型（MLLMs）在以人为中心的场景中感知和交互能力的全面基准，它关注对复杂人类意图的深刻理解和提供理性的、情境感知的反馈。
HumanSense is a comprehensive benchmark designed to evaluate the perception and interaction capabilities of Multimodal Large Language Models (MLLMs) in human-centered scenarios, focusing on deep understanding of complex human intentions and providing rational, context-aware feedback.
提供机构：
antgroup
