NeurIPS

Who Gets the Reward & Who Gets the Blame? Evaluation-Aligned Training Signals for Multi-LLM Agents

谁获得奖励 & 谁受到责备？面向多LLM智能体的评估对齐训练信号

Chih-Hsuan, Yang, Tanwi Mallick, Le Chen, Krishnan Raghavan, Amal Gueroudji, Ian T. Foster, Rajeev Thakur

机构 * Argonne National Laboratory（阿贡国家实验室） ； University of Chicago（芝加哥大学） 纠错

AI总结 提出一个理论框架，结合合作博弈归因与过程奖励建模，将系统级评估转化为智能体信用和消息级信号，用于多LLM智能体训练。

Comments Accepted at the NeurIPS 2025 Workshop on Bridging Language, Agent, and World Models for Reasoning and Planning (LAW 2025)