<a href=\"https://github.com/QwenLM/Qwen-AgentWorld\" rel=\"nofollow\">https://github.com/QwenLM/Qwen-AgentWorld</a></p>\n","updatedAt":"2026-06-24T03:25:13.466Z","author":{"_id":"622474f38dc6b0b64f5e903d","avatarUrl":"/avatars/d6b60a014277a8ec7d564163c5f644aa.svg","fullname":"Yuxin Zuo","name":"yuxinzuo","type":"user","isPro":false,"isHf":false,"isHfAdmin":false,"isMod":false,"followerCount":9,"isUserFollowing":false}},"numEdits":0,"identifiedLanguage":{"language":"en","probability":0.7100564241409302},"editors":["yuxinzuo"],"editorAvatarUrls":["/avatars/d6b60a014277a8ec7d564163c5f644aa.svg"],"reactions":[{"reaction":"🚀","users":["bambisheng"],"count":1}],"isReport":false}}],"primaryEmailConfirmed":false,"paper":{"id":"2606.24597","authors":[{"_id":"6a3b3b320a86ac3098d5d652","name":"Yuxin Zuo","hidden":false},{"_id":"6a3b3b320a86ac3098d5d653","name":"Zikai Xiao","hidden":false},{"_id":"6a3b3b320a86ac3098d5d654","name":"Li Sheng","hidden":false},{"_id":"6a3b3b320a86ac3098d5d655","name":"Fei Huang","hidden":false},{"_id":"6a3b3b320a86ac3098d5d656","name":"Jianhong Tu","hidden":false},{"_id":"6a3b3b320a86ac3098d5d657","name":"Yuxuan Liu","hidden":false},{"_id":"6a3b3b320a86ac3098d5d658","name":"Tianyi Tang","hidden":false},{"_id":"6a3b3b320a86ac3098d5d659","name":"Xiaomeng Hu","hidden":false},{"_id":"6a3b3b320a86ac3098d5d65a","name":"Yang Su","hidden":false},{"_id":"6a3b3b320a86ac3098d5d65b","name":"Qingfeng Lan","hidden":false},{"_id":"6a3b3b320a86ac3098d5d65c","name":"Yantao Liu","hidden":false},{"_id":"6a3b3b320a86ac3098d5d65d","name":"Qin Zhu","hidden":false},{"_id":"6a3b3b320a86ac3098d5d65e","name":"Yinger Zhang","hidden":false},{"_id":"6a3b3b320a86ac3098d5d65f","name":"Bowen Yu","hidden":false},{"_id":"6a3b3b320a86ac3098d5d660","name":"Haiquan Zhao","hidden":false},{"_id":"6a3b3b320a86ac3098d5d661","name":"Haiyang Xu","hidden":false},{"_id":"6a3b3b320a86ac3098d5d662","name":"Jianxin Yang","hidden":false},{"_id":"6a3b3b320a86ac3098d5d663","name":"Jiayang Cheng","hidden":false},{"_id":"6a3b3b320a86ac3098d5d664","name":"Junyang Wang","hidden":false},{"_id":"6a3b3b320a86ac3098d5d665","name":"Lianghao Deng","hidden":false},{"_id":"6a3b3b320a86ac3098d5d666","name":"Mingfeng Xue","hidden":false},{"_id":"6a3b3b320a86ac3098d5d667","name":"Tianyi Bai","hidden":false},{"_id":"6a3b3b320a86ac3098d5d668","name":"Yang Fan","hidden":false},{"_id":"6a3b3b320a86ac3098d5d669","name":"Yubo Ma","hidden":false},{"_id":"6a3b3b320a86ac3098d5d66a","name":"Yucheng Li","hidden":false},{"_id":"6a3b3b320a86ac3098d5d66b","name":"Zeyu Cui","hidden":false},{"_id":"6a3b3b320a86ac3098d5d66c","name":"Zhihai Wang","hidden":false},{"_id":"6a3b3b320a86ac3098d5d66d","name":"Zhihui Xie","hidden":false},{"_id":"6a3b3b320a86ac3098d5d66e","name":"Zhuorui Ye","hidden":false},{"_id":"6a3b3b320a86ac3098d5d66f","name":"An Yang","hidden":false},{"_id":"6a3b3b320a86ac3098d5d670","name":"Dayiheng Liu","hidden":false},{"_id":"6a3b3b320a86ac3098d5d671","name":"Jingren Zhou","hidden":false},{"_id":"6a3b3b320a86ac3098d5d672","name":"Ning Ding","hidden":false}],"publishedAt":"2026-06-23T00:00:00.000Z","submittedOnDailyAt":"2026-06-24T00:00:00.000Z","title":"Qwen-AgentWorld: Language World Models for General Agents","submittedOnDailyBy":{"_id":"6039478ab3ecf716b1a5fd4d","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/6039478ab3ecf716b1a5fd4d/_Thy4E7taiSYBLKxEKJbT.jpeg","isPro":true,"fullname":"taesiri","user":"taesiri","type":"user","name":"taesiri"},"summary":"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","upvotes":44,"discussionId":"6a3b3b330a86ac3098d5d673","ai_summary":"Language-based world models enable agentic environment simulation across multiple domains and enhance general agent performance through scalable simulation and improved downstream task performance.","ai_keywords":["world model","language models","agentic environment simulation","long chain-of-thought reasoning","foundation models","state transition dynamics","next-state-prediction reasoning","reinforcement learning","hybrid rubric-and-rule rewards","AgentWorldBench","agentic reinforcement learning","warm-up"],"ai_summary_model":"Qwen/Qwen2.5-Coder-32B-Instruct","organization":{"_id":"64c8b5837fe12ecd0a7e92eb","name":"Qwen","fullname":"Qwen","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6215ca5692c0ecfba9186921/hrRM50-6XcdWgg2AKpENG.jpeg"}},"canReadDatabase":false,"canManagePapers":false,"canSubmit":false,"hasHfLevelAccess":false,"upvoted":false,"upvoters":[{"_id":"620783f24e28382272337ba4","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/620783f24e28382272337ba4/zkUveQPNiDfYjgGhuFErj.jpeg","isPro":false,"fullname":"GuoLiangTang","user":"Tommy930","type":"user"},{"_id":"658c0b0574e79b9a8e9de89a","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/658c0b0574e79b9a8e9de89a/8HeMEOT5cLEzauXGlrqF_.jpeg","isPro":false,"fullname":"Xinping Zhao","user":"Yuki131","type":"user"},{"_id":"65377c30e48353201e6fdda0","avatarUrl":"/avatars/a8f803b6f2e598eaee9c52c0d2ddfc16.svg","isPro":false,"fullname":"Jiaheng Liu","user":"CheeryLJH","type":"user"},{"_id":"665ebae8bcbb98f60db0b4b1","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/665ebae8bcbb98f60db0b4b1/YTKM4qTZXh_2SeU8U7BfB.webp","isPro":false,"fullname":"Jiale Zhao","user":"Heisenburger2000","type":"user"},{"_id":"6523a521b0e0d574532b3b4b","avatarUrl":"/avatars/6b924802d08de02870dfeb31d90c66a5.svg","isPro":false,"fullname":"Liu","user":"Wenxuuuan","type":"user"},{"_id":"62f662bcc58915315c4eccea","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/62f662bcc58915315c4eccea/zOAQLONfMP88zr70sxHK-.jpeg","isPro":true,"fullname":"Yilun Zhao","user":"yilunzhao","type":"user"},{"_id":"622474f38dc6b0b64f5e903d","avatarUrl":"/avatars/d6b60a014277a8ec7d564163c5f644aa.svg","isPro":false,"fullname":"Yuxin Zuo","user":"yuxinzuo","type":"user"},{"_id":"646def60df618b303b419323","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/646def60df618b303b419323/JLJGYen4-5M8ivsLsSk0w.jpeg","isPro":false,"fullname":"Lei Wang","user":"demolei","type":"user"},{"_id":"643379416c6ecd58798421b3","avatarUrl":"/avatars/831db7eab2663abc33b176cf386b02f2.svg","isPro":false,"fullname":"Zhuoran Jin","user":"jinzhuoran","type":"user"},{"_id":"663f07d029be04778ba97871","avatarUrl":"/avatars/fb7c9d4a2c537d918a3267e7cbc03f04.svg","isPro":false,"fullname":"Xingtai Lv","user":"XingtaiHF","type":"user"},{"_id":"6898562e524e753b04240630","avatarUrl":"https://cdn-avatars.huggingface.co/v1/production/uploads/no-auth/ngRjX-dHx7SUNWavJ3IH6.png","isPro":false,"fullname":"Jincheng","user":"JinCheng777","type":"user"},{"_id":"6458e8ce4b7baff9a84aa0da","avatarUrl":"/avatars/c450f4885e68d28c22fd87f9efdfedec.svg","isPro":false,"fullname":"kaikai zhao","user":"LifeIsSoSolong","type":"user"}],"acceptLanguages":["en"],"dailyPaperRank":1,"organization":{"_id":"64c8b5837fe12ecd0a7e92eb","name":"Qwen","fullname":"Qwen","avatar":"https://cdn-avatars.huggingface.co/v1/production/uploads/6215ca5692c0ecfba9186921/hrRM50-6XcdWgg2AKpENG.jpeg"},"markdownContentUrl":"https://huggingface.co/buckets/huggingchat/papers-content/resolve/2606/2606.24597.md","query":{}}">

Get this paper in your agent:

Don't have the latest CLI?

Cite arxiv.org/abs/2606.24597 in a Space README.md to link it from this page.