Vector Institute· Jennifer Jenkins·· 2025-08-09AI 评分38
HumaniBench：首个以人为中心的多模态大模型评测基准
When AI Meets Human Matters: Evaluating Multimodal Models Through a Human-Centred Lens – Introducing HumaniBench
AI 导读
summary_zh: HumaniBench 是首个针对大多模态模型（LMMs）以人为中心的综合评测基准，涵盖公平性、伦理、理解、推理、语言包容性、同理心与鲁棒性七项原则。
来源：Vector Institute · vectorinstitute.ai