世界模型论文趋势 2023–2026

主要结论

• 世界模型论文从 2023 年的 23 篇增长到 2026 年的 250 篇，三年间扩大到约 11 倍。
• NeurIPS 是发世界模型论文最多的会议，2023–2026 年共 159 篇，占四个会议总量的 37%。
• 交叉方向里涨得最快的是多模态大模型：同时属于两个主题的论文从 2023 年的 1 篇增至 2026 年的 34 篇。
• 在「前沿系统」的 4 个主题里，它的 2026 年论文数排第 4；同组增长最快的AI 智能体是 12 倍，世界模型是 11 倍。

代表论文

按年份和会议均匀挑选，方便快速看到每一届的样貌

• NeurIPS 2026A Hierarchical World Model for Driving
• ICML 2026A Deep Learning Model of Mental Rotation Informed by Interactive VR Experiments
• ICLR 2026Astra: General Interactive World Model with Autoregressive Denoising
• CVPR 2026Active Intelligence in Video Avatars via Closed-loop World Modeling
• NeurIPS 2025COME: Adding Scene-Centric Forecasting Control to Occupancy World Model
• ICML 2025A Causal World Model Underlying Next Token Prediction: Exploring GPT in a Controlled Environment
• ICLR 2025AdaWM: Adaptive World Model based Planning for Autonomous Driving
• CVPR 2025DIO: Decomposable Implicit 4D Occupancy-Flow World Model
• NeurIPS 2024A Data-Centric Perspective on Evaluating Machine Learning Models for Tabular Data
• ICML 20243D-VLA: A 3D Vision-Language-Action Generative World Model
• ICLR 2024Copilot4D: Learning Unsupervised World Models for Autonomous Driving via Discrete Diffusion
• CVPR 2024DriveWorld: 4D Pre-trained Scene Understanding via World Models for Autonomous Driving
• NeurIPS 2023Action Inference by Maximising Evidence: Zero-Shot Imitation from Observation with World Models
• ICML 2023Do Embodied Agents Dream of Pixelated Sheep: Embodied Decision Making using Language Guided World Modelling
• ICLR 2023Dynamic Update-to-Data Ratio: Minimizing World Model Overfitting
• NeurIPS 2026A World Model of Radiologist Reading for Medical Image Representation Learning
• ICML 2026Behavior-Invariant Task Representation Learning with Transformer-based World Models for Offline Meta-Reinforcement Learning
• ICLR 2026Building spatial world models from sparse transitional episodic memories
• CVPR 2026Chain of World World Model Thinking in Latent Motion
• NeurIPS 2025DMWM: Dual-Mind World Model with Long-Term Imagination

全部世界模型论文

共 106 篇，第 1 / 3 页

1. CVPR 2025CheXWorld: Exploring Image World Modeling for Radiograph Representation Learning
2. CVPR 2025DIO: Decomposable Implicit 4D Occupancy-Flow World Model
3. CVPR 2025DriveDreamer4D: World Models Are Effective Data Machines for 4D Driving Scene Representation
4. CVPR 2025EchoWorld: Learning Motion-Aware World Models for Echocardiography Probe Guidance
5. CVPR 2025Exploration-Driven Generative Interactive Environments
6. CVPR 2025GaussianWorld: Gaussian World Model for Streaming 3D Occupancy Prediction
7. CVPR 2025GEM: A Generalizable Ego-Vision Multimodal World Model for Fine-Grained Ego-Motion, Object Dynamics, and Scene Composition Control
8. CVPR 2025MaskGWM: A Generalizable Driving World Model with Video Mask Reconstruction
9. CVPR 2025Missing Target-Relevant Information Prediction with World Model for Accurate Zero-Shot Composed Image Retrieval
10. CVPR 2025Navigation World Models
11. CVPR 2025Neural Motion Simulator Pushing the Limit of World Models in Reinforcement Learning
12. CVPR 2025ReconDreamer: Crafting World Models for Driving Scene Reconstruction via Online Restoration
13. CVPR 2025SceneDiffuser++: City-Scale Traffic Simulation via a Generative World Model
14. ICLR 2025AdaWM: Adaptive World Model based Planning for Autonomous Driving
15. ICLR 2025COMBO: Compositional World Models for Embodied Multi-Agent Cooperation
16. ICLR 2025Discrete Codebook World Models for Continuous Control
17. ICLR 2025Drama: Mamba-Enabled Model-Based Reinforcement Learning Is Sample and Parameter Efficient
18. ICLR 2025Dream to Manipulate: Compositional World Models Empowering Robot Imitation Learning with Imagination
19. ICLR 2025Dreamweaver: Learning Compositional World Models from Pixels
20. ICLR 2025Efficient Exploration and Discriminative World Model Learning with an Object-Centric Abstraction
21. ICLR 2025Empowering LLM Agents with Zero-Shot Optimal Decision-Making through Q-learning
22. ICLR 2025Enhancing End-to-End Autonomous Driving with Latent World Model
23. ICLR 2025FLIP: Flow-Centric Generative Planning as General-Purpose Manipulation World Model
24. ICLR 2025FOSP: Fine-tuning Offline Safe Policy through World Models
25. ICLR 2025Hierarchical World Models as Visual Whole-Body Humanoid Controllers
26. ICLR 2025Language Agents Meet Causality -- Bridging LLMs and Causal World Models
27. ICLR 2025Learning Transformer-based World Models with Contrastive Predictive Coding
28. ICLR 2025Learning View-invariant World Models for Visual Robotic Manipulation
29. ICLR 2025MMWorld: Towards Multi-discipline Multi-faceted World Model Evaluation in Videos
30. ICLR 2025Open-World Reinforcement Learning over Long Short-Term Imagination
31. ICLR 2025PWM: Policy Learning with Multi-Task World Models
32. ICLR 2025Scaling Offline Model-Based RL via Jointly-Optimized World-Action Model Pretraining
33. ICLR 2025Semi-Supervised Vision-Centric 3D Occupancy World Model for Autonomous Driving
34. ICLR 2025Simple, Good, Fast: Self-Supervised World Models Free of Baggage
35. ICLR 2025VisualPredicator: Learning Abstract World Models with Neuro-Symbolic Predicates for Robot Planning
36. ICLR 2025Web Agents with World Models: Learning and Leveraging Environment Dynamics in Web Navigation
37. ICLR 2025World Model on Million-Length Video And Language With Blockwise RingAttention
38. ICML 2025A Causal World Model Underlying Next Token Prediction: Exploring GPT in a Controlled Environment
39. ICML 2025Accurate and Efficient World Modeling with Masked Latent Transformers
40. ICML 2025AdaWorld: Learning Adaptable World Models with Latent Actions
41. ICML 2025Continual Reinforcement Learning by Planning with Online World Models
42. ICML 2025DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning
43. ICML 2025Empowering World Models with Reflection for Embodied Video Prediction
44. ICML 2025FOUNDER: Grounding Foundation Models in World Models for Open-Ended Embodied Decision Making
45. ICML 2025General agents need world models
46. ICML 2025Graph World Model
47. ICML 2025How Far Is Video Generation from World Model: A Physical Law Perspective
48. ICML 2025Improving Transformer World Models for Data-Efficient RL
49. ICML 2025Multi-Stage Manipulation with Demonstration-Augmented Reward, Policy, and World Model Learning
50. ICML 2025Neurosymbolic World Models for Sequential Decision Making

常见问题

2023–2026 年四大会议一共有多少篇世界模型论文？

共 425 篇，按年份分别是 2023 年 23 篇、2024 年 46 篇、2025 年 106 篇、2026 年 250 篇。

NeurIPS 2026 有多少篇世界模型论文？

NeurIPS 2026 有 85 篇世界模型论文，上一届是 42 篇。

ICLR 2026 有多少篇世界模型论文？

ICLR 2026 有 56 篇世界模型论文，上一届是 24 篇。

ICML 2026 有多少篇世界模型论文？

ICML 2026 有 69 篇世界模型论文，上一届是 27 篇。

CVPR 2026 有多少篇世界模型论文？

CVPR 2026 有 40 篇世界模型论文，上一届是 13 篇。

哪个会议的世界模型论文最多？

NeurIPS，2023–2026 年共 159 篇。