Interactive World Models

Follow this topic Ask about this topic

Momentum

13 papers in the last four weeks, up 30% on the four weeks before. 0.1% of all new papers.

Jul 13Week of Sep 28

Latest papers 86

All topics

WorldCycle: Self-Verifiable Reinforcement Learning for Long-Horizon Video World Models
Aug 5, 2026Bohai Gu, Yueyang Yuan, Taiyi Wu +9World Model-Based Planning Action-Conditioned World Models
1The Hong Kong University of Science and Technology · 2Wuhan University · 3AI Technology Center, Tencent Video, Tencent
Reconstructing Persistent Worlds from Narratives for Narrative-Grounded Interactive Experiences
Aug 3, 2026Yi-Chun ChenInteractive Storytelling Interactive World Models
National Cheng Kung University Tainan City 701, Taiwan
Quo Vadis, World Modeling?
Aug 3, 2026Yu Yang, Xuemeng Yang, Licheng Wen +17World Model Learning Lifelong Learning Agents
WorldExam: Benchmarking World Models from Apparent Appearance to Inherent Reactivity
Aug 3, 2026Yuxue Yang, Shuyao Shang, Jiahe Wang +13Controllable Video Generation World Model Evaluation
1CASIA · 4AMAP · 5THU +2
WM-Cov: Test Adequacy for Interactive World-Model-Style Autonomous Driving Simulation
Jul 31, 2026Jianxun Cui, Ping Wu, Stanisa Peric +2Autonomous Driving Safety Evaluation World Model Evaluation
School of Transportation Science and Engineering, Harbin Institute of Technology, Harbin, China · Chongqing Research Institute of HIT, Chongqing, China · Chongqing Changan Automobile Co., Ltd., Chongqing, China +2
ABot-World-0: Infinite Interactive World Rollout on a Single Desktop GPU
Jul 21, 2026Fan Jiang, Zhaoxu Sun, Mengchao Wang +38Action-Conditioned Video Generation Long-Horizon Video Generation
Generative World Renderer at the Speed of Play
Jul 21, 2026Guixu Lin, Zheng-Hui Huang, Siqi Yang +3Real-Time Interactive Video Generation Streaming Video Generation
1Alaya Lab · University of California, Merced
AlayaWorld: Interactive Long-Horizon World Modeling -- Full Technical Report
Jul 20, 2026AlayaWorld Team, Kaipeng Zhang, Chuanhao Li +15Video Diffusion Models Real-Time Interactive Video Generation
Alaya Lab
EvolvingWorld: An Open-Schema Framework for Co-Evolving Role-Play Agents and World Model in Interactive Literary World
Jul 19, 2026Qing Zong, Yue Guo, Mengxin Yang +2Long-Horizon Agent Evaluation LLM Agents
Hong Kong University of Science and Technology · LIGHTSPEED · Huazhong University of Science and Technology +1
From Pixels to States: Rethinking Interactive World Models as Game Engines
Jul 15, 2026Zhen Li, Zian Meng, Shuwei Shi +4Action-Conditioned Video Generation Action-Conditioned World Models
1Alaya Lab
Stateful Worlds, Stateless Elasticity: Exact-State Serving for Interactive World Models
Jul 11, 2026Jin Li, Jiawei ChenDistributed Inference Online Resource Allocation
Harvard University · Harvard University Cambridge, MA, USA · Independent Researcher
Infinite Worlds with Versatile Interactions
Jul 8, 2026Zelin Gao, Qiuyu Wang, Jiapeng Zhu +17Real-Time Interactive Video Generation World Models
AlayaWorld: Long-Horizon and Playable Video World Generation
Jul 7, 2026AlayaWorld Team, Kaipeng Zhang, Chuanhao Li +14Real-Time Interactive Video Generation Long-Horizon Video Generation
Alaya Lab
MoWorld: A Flash World Model
Jul 7, 2026Team Moxin, Deyi Ji, Tianrun Chen +26Efficient Inference Interactive World Models
MoWorld Team, Moxin Technology
Multiplayer Interactive World Models with Representation Autoencoders
Jul 6, 2026Anthony Hu, Václav Volhejn, Adrien Ramanana Rahary +24Action-Conditioned World Models Latent World Models
1General Intuition · 2Kyutai · ‡École nationale des ponts et chaussées +1
OPINE-World: Programmatic World Modeling with Ontology-error-Prioritized Interactive Exploration for ARC-AGI-3
Jul 1, 2026David Courtis, Wenhao Li, Scott SannerWorld Model Learning World Models
Department of Computer Science, University of Toronto
WorldRoamBench: An Open-World Benchmark for Long-Horizon Stability of Interactive World Models
Jun 30, 2026Ting-Bing Xu, Jiacheng Sui, Zhe Gao +12World Model Evaluation Interactive World Models
AMAP CV Lab, Alibaba Group · Nanjing University · Tsinghua University +1
DreamForge-World 0.1 Preview: A Low-Compute Real-Time Controllable World Model
Jun 29, 2026Daniyel Ayupov, Artur Markov-TsoyReal-Time Interactive Video Generation World Models
DreamForge AI Lab, Kazakhstan
Walking in the Implicit: Interactive World Exploration via Neural Scene Representation
Jun 29, 2026Zhiqi Li, Chengrui Dong, Zhenhua Du +6Camera-Conditioned Video Generation Long-Horizon Video Generation
Zhejiang University · Westlake University · Afari Intelligent Drive
IOI: Decoupling Kinematics and Physics for Interactive World Models
Jun 22, 2026Chengyu Bai, Peidong Jia, Tiecheng Guo +11Action-Conditioned Video Generation World Models for Robotics
1Beijing Innovation Center of Humanoid Robotics · 2Peking University
EgoCS-400K: An Egocentric Gameplay Dataset for World Models
Jun 16, 2026Rongjin Guo, Dong Liang, Yuhao Liu +4World Model Learning Egocentric Video Understanding
City University of Hong Kong
ActWorld: From Explorable to Interactive World Model via Action-Aware Memory
Jun 16, 2026Zhexiao Xiong, Yizhi Song, Hao Kang +11HOI Generation World Model Learning
1Washington University in St. Louis · 2Intelligent Creation, ByteDance · *Work done during internship at ByteDance
DreamX-World 1.0: A General-Purpose Interactive World Model
Jun 15, 2026DreamX Team, Yancheng Bai, Rui Chen +20Camera-Controlled Video Generation Real-Time Interactive Video Generation
DreamX Team
Orchestrated Reality: From Role-Play to Living, Playable Game Worlds -- LLM-Driven World Simulation as a Parameterized-Action POMDP
Jun 14, 2026Yuhang Huang, Chenmiao Li, Chaowei FangLarge Language Model-Based Role-Play Simulation LLM Agent Orchestration
The University of Tokyo, Japan · Individual Researcher, Japan
MoVerse: Real-Time Video World Modeling with Panoramic Gaussian Scaffold
Jun 11, 2026Yang Zhou, Ziheng Wang, Yuqin Lu +43D Gaussian Splatting Camera-Conditioned Video Generation
South China University of Technology · Orange Team, Moku Lab, HUJING Digital Media & Entertainment Group · Columbia University +1
BiWM: Advancing Open-Source Interactive Video World Models with Bidirectional Autoregression
Jun 8, 2026Shaohao Rui, Xiaofeng Mao, Zhanyu Zhang +5Action-Conditioned Video Generation Camera-Controlled Video Generation
LynnReal AI · Shanghai Innovation Institute · Shanghai Jiao Tong University +1
$ω$ -EVA: Envision, Verify, and Act with Latent Interactive World Models
Jun 8, 2026Zhenguo Sun, Yu Sun, Hande Huang +1Robot Policy Learning World Models for Robotics
AnchorWorld: Embodied Egocentric World Simulation with View-based Evolution Customization
Jun 5, 2026Yu Li, Menghan Xia, Gongye Liu +8World Models for Robotics World Models
1Tsinghua University · ∗This work was conducted during the author’s internship at Kling Team, Kuaishou Technology. · 2HUST +3
From Zero to Hero: Training-Free Custom Concept Spawning in World Models
Jun 1, 2026Kiymet Akdemir, Pinar YanardagWorld Models Autoregressive Video Generation
Virginia Tech
Towards Interactive Video World Modeling: Frontiers, Challenges, Benchmarks, and Future Trends
May 31, 2026Jiuming Liu, Chaojun Ni, Mengmeng Liu +7Action-Conditioned Video Generation Long-Horizon Video Generation
Department of Engineering, University of Cambridge, U.K. · Peking University, China · University of Twente, Netherlands +4

Aug 5, 2026, cs.AIJ/K move · Enter open · S save

WorldCycle: Self-Verifiable Reinforcement Learning for Long-Horizon Video World Models

Bohai Gu, Yueyang Yuan, Taiyi Wu, Dazhao Du+8

1The Hong Kong University of Science and Technology · 2Wuhan University · 3AI Technology Center, Tencent Video, Tencent

World Model-Based Planning Action-Conditioned World Models Reinforcement Learning with Verifiable Rewards World Model Evaluation Interactive World Models Video World Models

Open paper AskPDF arXiv

Interactive video world models are essential for long-horizon planning and exploration, yet they suffer from compounding errors. Post-training methods such as reinforcement learning (RL) can improve these models, but they hit a verification bottleneck: for arbitrary action sequences, no ground-truth future state exists to measure long-term drift. Our key insight is that reversible action cycles make this verification possible: a sequence composed with its inverse must analytically return to the initial state, yielding annotation-free supervision on long-horizon correctness. Building on this, we introduce WorldCycle, a self-verifiable RL framework that constructs closed action cycles and their repeated executions from ordinary action sequences, and optimizes two complementary rewards: a spatial closure reward enforcing symmetry between mirrored forward and reverse segments, and a temporal consistency reward aligning states across repeated cycle executions. These rewards force the model to learn actions as consistent state operators rather than memorized temporal patterns, and extend naturally to out-of-distribution composite action cycles that the base model handles poorly. We further release CycleBench, a diagnostic benchmark for state-returning ability under complex action structures. WorldCycle reduces state returning drift by up to 44% and lifts composite-action accuracy nearly 4x over the base model, providing a vital foundation for physically grounded world models.

Momentum

Related topics

Latest papers 86