arXiv cs.AI· Haochen Liu, Qianhao Zhu, Zixuan Wang, Dingming Li, Tongbo Chen, Peng Qiu, Yingqi Qu, Weiming Lu, Jun Xiao, Yongliang Shen·· 1 天前AI 评分39
Actor、Judge 还是 Teacher？理解长程智能体中的决策模型
Actor, Judge, or Teacher? Understanding Decision Models in Long-Horizon Agents
AI 导读
研究比较三个决策模型与三个通用 LLM 在 ALFWorld 上作为 actor 和 judge 的表现，并将 JEV 作为 teacher 用于 ALFWorld 和 WebShop。
正文 · 原文
Abstract:Decision models answer a question with a typed output, such as a choice among options with a probability for each, so one model can choose an agent's actions, evaluate actions already taken, or score every action to teach a policy with reinforcement learning. Prior work has studied them mainly as actors, judging their usefulness by how well they act. We ask whether a decision model that acts poorly can still teach well. We compare three decision models and three general LLMs as actors and judges on ALFWorld, and use JEV as a teacher on ALFWorld and WebShop. As actors, all six models are pulled away from a subgoal that must come first by an easier one. As judges that see the whole trajectory, the same models credit the steps toward that subgoal, and asking whether each action helped, rather than whether the task will succeed, separates useful from useless actions. As a teacher, JEV raises a Qwen2.5-1.5B policy over GRPO from 72.8% to 82.9% success on ALFWorld and from 56.8% to 71.4% on WebShop. These results suggest that a poor actor can still teach well, because judging an action after the episode ends reveals the consequences that acting must anticipate. Code is available at this https URL.
| Subjects: | Artificial Intelligence (cs.AI); Computation and Language (cs.CL) |
| Cite as: | arXiv:2611.17167 [cs.AI] |
| (or arXiv:2611.17167v1 [cs.AI] for this version) | |
| https://doi.org/10.48550/arXiv.2611.17167 arXiv-issued DOI via DataCite (pending registration) |
Submission history
From: Zixuan Wang [view email]
[v1]
Thu, 12 Nov 2026 17:42:46 UTC (5,255 KB)
来源：arXiv cs.AI · arxiv.org