研究人员推出了PIVOTSBench，这是一个旨在评估多模态大语言模型（MLLM）理解和推理人际关系能力的新基准。该基准源自Social-IQ 2.0和YouTube数据，包含评估模型预测关系维度和识别关键视觉线索能力的任务。评估涵盖了专有和开源的MLLM，研究探讨了视觉模态和对话上下文的影响。 AI
影响 该基准有望推动具备改进的社交推理能力MLLM的发展，这对于更自然的人机交互至关重要。
排序理由 该集群描述了一篇介绍AI模型评估基准的新学术论文。
在 Hugging Face Daily Papers 阅读 →
AI 生成摘要 · Google Gemini · 来自 2 个来源。 我们如何撰写摘要 →
完整方法见我们的编辑标准。
Humans possess an innate ability to understand fine-grained interpersonal relationships, which is central to everyday social interactions. Although such reasoning is inherently multimodal, it remains largely unexplored by existing multimodal large language models (MLLMs). To addr…
Humans possess an innate ability to understand fine-grained interpersonal relationships, which is central to everyday social interactions. Although such reasoning is inherently multimodal, it remains largely unexplored by existing multimodal large language models (MLLMs). To addr…