AI 中文总结
本研究提出以智能体为中心的交互式世界代理，将世界建模范式从物理状态转换转向智能体可用信息转换，划分六种代理类型与三个赋能层面，为构建赋能智能体的世界代理建立路线图。
AI 中文摘要
持续改进的智能体需要超越静态监督的动态交互反馈，但直接在真实环境中交互成本高、速度慢、不安全且难以并行化。世界建模提供了一种自然的中间代理，允许智能体在执行真实动作前查询成本更低、可控性更强的反馈。经典世界模型主要通过未来物理状态预测来实例化这种代理，该表述虽有用但对于需要超越原始状态转换的可操作反馈的智能体而言过于狭窄。本研究提出了以智能体为中心的交互式世界代理，将基本范式从物理状态转换转变为智能体可用的信息转换，例如执行结果、检索到的经验或技能以及验证信号，从而拓宽世界建模的范围，为持续改进的智能体提供多样化的反馈。为系统地描绘该设计空间，我们根据反馈模态将世界代理组织为六种功能形式：动态代理、空间代理、执行代理、记忆/经验代理、技能代理以及奖励/验证代理，共同刻画了世界建模服务于智能体改进的主要方式。我们进一步分析这些代理如何在三个递进层面赋能智能体：L.1 推理时指导，即代理输出丰富上下文信息以实现更优决策；L.2 训练时优化，即代理输出为策略学习提供奖励、批评或合成回滚；L.3 智能体-代理协同进化，即真实环境证据持续更新代理与智能体以实现协同进化。最终，本研究将世界建模重塑为以智能体为中心的范式，为构建赋能智能体更好规划、更快学习和持续进化的世界代理建立了路线图。
英文摘要
Continually improving agents require dynamic interaction feedback beyond static supervision, yet direct real-environment interaction is costly, slow, unsafe, and hard to parallelize. World modeling offers a natural intermediate proxy that allows agents to query lower-cost, more controllable feedback before committing to real actions. Classical world models instantiate this proxy primarily through future physical-state prediction, a formulation useful yet narrow for agents that require actionable feedback beyond raw state transitions. In this work, we conceptualize Agent-Centric Interactive World Proxies, shifting the fundamental paradigm from physical state transitions to agent-usable information transitions, such as execution outcomes, retrieved experiences or skills, and verification signals, broadening the scope of world modeling to provide versatile feedback for continually improving agents. To systematically map this design space, we organize world proxies into six functional forms based on their feedback modalities: dynamics, spatial, execution, memory/experience, skill, and reward/verification proxies, which together characterize the primary ways world modeling serves agent improvement. We further analyze how these proxies empower agents across three progressive levels: L.1 Inference-Time Guidance, where proxy outputs enrich in-context information for superior decisions; L.2 Training-Time Optimization, where proxy outputs yield rewards, critiques, or synthetic rollouts for policy learning; and L.3 Agent-Proxy Co-Evolution, where real-environment evidence continuously updates both the proxy and the agent for co-evolution. Ultimately, this work recasts world modeling into an agent-centric paradigm, establishing a roadmap for building world proxies that empower agents to plan better, learn faster, and evolve continually.
CommentsTechnical Blog at https://worldbench.github.io/awesome-agentic-world-model GitHub Repo at https://github.com/worldbench/awesome-agentic-world-model
