LLM规范性能力评估框架提出

1. 12:00
   arXiv · Artificial Intelligence

   LLM 规范性能力的基础、设计与挑战

   论文提出规范性能力（normative competence）概念，即 AI 智能体仅凭交互识别社区所执行规范的能力，并构建多智能体社区辩论场景加以隔离评估。基线 LLM 智能体即便遵守规范能提升准确率仍学不会；规范性模块对规范风格与底层模型高度敏感，且会把非规范的特异行为一并模仿。该工作自称首次对 LLM 规范性能力进行可操作化评测。

Current heat 9·Comparable peak 10（Oct 9）·Comparable change over 24 hours –

The trend compares only the same participants observed continuously; its range may be smaller than the current heat count. Move or click on the chart to inspect hourly heat; use the left and right arrow keys to switch.