HappyWorld-Bench

  • 类型:arxiv
  • 标识:2609.24308
  • 链接:https://arxiv.org/abs/2609.24308
  • 主分类:evaluation
  • 形态:benchmark
  • TLDR:Evaluating world models requires assessing both the quality of the worlds they generate and their consistency and responsiveness under exploration, interaction, and modification. We introduce HappyWorld-Bench, a comprehensive benchmark that evaluates whether generated worlds remain reliable as agents interact with them. Our design is built on a hierarchical capability framework of six world capabilities (W1-W6), from generative construction to unified world modeling, instantiated across three independent evaluation tracks: video world models, spatial world models, and embodied world models. Ha
  • 待LLM分类:否
  • 来源文件
  • /inbox/tom/_candidates/2026-09-24-agent-rag-longcontext-candidates.json