用大模型模拟真实世界，让智能体更聪明地决策。
Qwen-AgentWorld: Language World Models for General Agents
- 用7个领域的1000万条交互数据训练语言世界模型，支持长链推理。
- 在9个基准上超越现有模型，提升智能体在真实环境中的表现。
- 既可当仿真器，也可做预训练基础模型，适合强化学习与通用智能体研究。
世界模型基于当前观测与动作预测环境动态，是智能体推理与规划的核心机制。本文探索语言模型驱动的世界建模如何推动通用智能体边界。首先，构建了首个能覆盖7个领域、支持长链推理的语言世界模型——Qwen-AgentWorld-35B-A3B与Qwen-AgentWorld-397B-A17B。基于超过1000万条真实环境交互轨迹，通过三阶段训练：CPT从状态转移动态与专业语料中注入通用建模能力，SFT激活下一步状态预测推理，RL则通过混合评分与规则奖励框架优化仿真保真度。为评估语言世界模型，提出AgentWorldBench，基于5个前沿模型在9个标准基准上的真实交互构建。实验表明，Qwen-AgentWorld显著优于现有先进模型。其次，探索两种互补范式：作为解耦环境仿真器，可高效生成数千个真实环境，提升强化学习性能；作为统一智能体基础模型，世界模型训练可有效预热，提升7个任务上的下游表现。代码已开源。
原文摘要 · Abstract (English)
A world model predicts environment dynamics based on current observations and actions, serving as a core cognitive mechanism for reasoning and planning. In this work, we investigate how world modeling based on language models can further push the boundaries of general agents. (i) We first focus on building foundation models for agentic environment simulation. We introduce Qwen-AgentWorld-35B-A3B and Qwen-AgentWorld-397B-A17B, the first language world models capable of simulating agentic environments covering 7 domains via long chain-of-thought reasoning. Leveraging more than 10M environment interaction trajectories of 7 domains in real-world environments, we develop Qwen-AgentWorld through a three-stage training pipeline: CPT injects general-purpose world modeling capabilities from the state transition dynamics and augmented professional corpora, SFT activates next-state-prediction reasoning, and RL sharpens simulation fidelity through a tailored framework with hybrid rubric-and-rule rewards. To evaluate language world models, we present AgentWorldBench, a comprehensive benchmark constructed from real-world interactions of 5 frontier models on 9 established benchmarks. Empirical results demonstrate that Qwen-AgentWorld significantly outperforms existing frontier models. (ii) Beyond foundation models, we further investigate two complementary paradigms through which world modeling enhances general agents. First, as a decoupled environment simulator, Qwen-AgentWorld supports scalable and controllable simulation of thousands of real-world environments for agentic RL, yielding gains that surpass real-environment training alone. Second, as a unified agent foundation model, world-model training acts as a highly effective warm-up that improves downstream performance across 7 agentic benchmarks. Code: https://github.com/QwenLM/Qwen-AgentWorld
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。