Agent-Editing World Model：重新思考 LLM 智能体的世界建模
原标题：Paper page - Agent-Editing World Model: Rethinking World Modeling for LLM Agents
AI 摘要
该论文提出 Agent-Editing World Model（AEWM），重新思考长时程 LLM 智能体的世界建模方式，将目标从预测环境观测转向建模推理与动作如何影响未来任务进展。AEWM 结合 Action Judge（区分关键、探索性、噪声决策）与 State Revision（直接编辑噪声推理-动作延续），其推理框架 EditAct 与真实环境交互结合。在 Action Judge 基准上达到 70.5% macro-F1，超过最强前沿基线 10.6 个百分点；在六个基准和三个智能体骨干上，EditAct 平均提升 3.2-6.7 分。
正文节选
Agent-Editing World Model: Rethinking World Modeling for LLM Agents Abstract Recent advances in large language models (LLMs) have enabled agents to tackle long-horizon tasks across diverse environments. To further improve agent performance, existing language world models typically predict environment observations, yet reconstructing high-entropy, execution-dependent tool responses offers limited value when real feedback is available. Meanwhile, agents suffer from task-state contamination, where