Arena 提出以交互作为评判新范式

原标题:Arena research intern and PhD candidate at Carnegie Mellon University, Wayne Chi, explains Interacti...
58AI 研判分

Arena.ai 研究实习生、卡内基梅隆大学博士生 Wayne Chi 提出“交互即评判”(Interaction-as-a-Judge)方法,让计算机使用智能体在真实应用与游戏中操作,生成交互轨迹供 LLM 评判器或训练好的奖励模型评估。该框架覆盖两类评估:基于 Code Arena 真实对战的偏好评估,以及基于 Agent Arena 用户投诉挖掘需求的正确性评估。

报道正文

Arena.ai (@arena) · 2026-11-12 21

Arena research intern and PhD candidate at Carnegie Mellon University, Wayne Chi, explains Interaction-as-a-Judge: letting computer-use agents play games and use apps to better improve evaluation and downstream generation.

A computer-use agent navigates an application, producing interaction traces for evaluation by an LLM judge or a trained reward model. The talk covers preference evaluation using real Code Arena battles and correctness evaluation using requirements mined from real user complaints in Agent Arena.

Watch the full talk: