MM-Hallu/MME
收藏官方服务:
资源简介:
MME(多模态评估)是一个全面的基准,用于评估多模态大语言模型在多种感知和认知任务上的表现。
MME (MultiModal Evaluation) is a comprehensive benchmark for evaluating multimodal large language models across diverse perception and cognition tasks.
提供机构:
MM-Hallu
MME(多模态评估)是一个全面的基准,用于评估多模态大语言模型在多种感知和认知任务上的表现。
MME (MultiModal Evaluation) is a comprehensive benchmark for evaluating multimodal large language models across diverse perception and cognition tasks.