Original Summary

A benchmark for epistemic arbitration and misinformation resistance in AI agents. Evaluates an agent's ability to distinguish independent evidence from coordinated repetition and manufactured consensus within synthetic social environments.

  • 情报分类：技术学习与提效
  • 分类依据：内容涉及技术、AI、软件工具或工程实践
  • 信息来源：GitHub · AI 新项目
  • 发布时间：2026/9/21 20:36:41