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 111 篇，第 2 / 3 页

51. ICML 2026SC$^{2}$-WM: A Self-Correcting World Model with Closed-Loop Feedback for Vision-and-Language Navigation in Continuous Environments
52. ICML 2026Scaling Real-World Robot Policy Evaluation via Discrete Diffusion World Model
53. ICML 2026Self-supervised Hierarchical Visual Reasoning with World Model
54. ICML 2026Structure Abstraction and Generalization in a Hippocampal-Entorhinal Inspired World Model
55. ICML 2026TaskLoom: Weaving Knowledge Across Tasks in World Models
56. ICML 2026TimeSpot: Benchmarking Geo-Temporal Understanding in Vision–Language Models in Real-World Settings
57. ICML 2026Towards Practical World Model-based Reinforcement Learning for Vision-Language-Action Models
58. ICML 2026Uncertainty-Guided Exploration and Stable Planning for Sparse-Reward Manipulation from Limited Demonstrations
59. ICML 2026VectorWorld: Efficient Streaming World Model via Diffusion Flow on Vector Graphs
60. ICML 2026VJEPA: Variational Joint Embedding Predictive Architectures as Probabilistic World Models
61. ICML 2026VLAW: Iterative Co-Improvement of Vision-Language-Action Policy and World Model
62. ICML 2026WebWorld: A Large-Scale World Model for Web Agent Training
63. ICML 2026WestWorld: A Knowledge-Encoded Scalable Trajectory World Model for Diverse Robotic Systems
64. ICML 2026What You Think is What You See: Driving Exploration in VLM Agents via Visual-Linguistic Curiosity
65. ICML 2026World Guidance: World Modeling in Condition Space for Action Generation
66. ICML 2026World Models in Pieces: Structural Certification of General Agents
67. ICML 2026WorldCache: Accelerating World Models for Free via Heterogeneous Token Caching
68. ICML 2026WorldCompass: Reinforcement Learning for Long-Horizon World Models
69. ICML 2026WorldPlay: Towards Long-Term Geometric Consistency for Real-Time Interactive World Modeling
70. ICML 2025A Causal World Model Underlying Next Token Prediction: Exploring GPT in a Controlled Environment
71. ICML 2025Accurate and Efficient World Modeling with Masked Latent Transformers
72. ICML 2025AdaWorld: Learning Adaptable World Models with Latent Actions
73. ICML 2025Continual Reinforcement Learning by Planning with Online World Models
74. ICML 2025DINO-WM: World Models on Pre-trained Visual Features enable Zero-shot Planning
75. ICML 2025Empowering World Models with Reflection for Embodied Video Prediction
76. ICML 2025FOUNDER: Grounding Foundation Models in World Models for Open-Ended Embodied Decision Making
77. ICML 2025General agents need world models
78. ICML 2025Graph World Model
79. ICML 2025How Far Is Video Generation from World Model: A Physical Law Perspective
80. ICML 2025Improving Transformer World Models for Data-Efficient RL
81. ICML 2025Multi-Stage Manipulation with Demonstration-Augmented Reward, Policy, and World Model Learning
82. ICML 2025Neurosymbolic World Models for Sequential Decision Making
83. ICML 2025Novelty Detection in Reinforcement Learning with World Models
84. ICML 2025PIGDreamer: Privileged Information Guided World Models for Safe Partially Observable Reinforcement Learning
85. ICML 2025PISA Experiments: Exploring Physics Post-Training for Video Diffusion Models by Watching Stuff Drop
86. ICML 2025PlaySlot: Learning Inverse Latent Dynamics for Controllable Object-Centric Video Prediction and Planning
87. ICML 2025Reward-free World Models for Online Imitation Learning
88. ICML 2025Scaling Laws for Pre-training Agents and World Models
89. ICML 2025SENSEI: Semantic Exploration Guided by Foundation Models to Learn Versatile World Models
90. ICML 2025SOLD: Slot Object-Centric Latent Dynamics Models for Relational Manipulation Learning from Pixels
91. ICML 2025Time-Aware World Model for Adaptive Prediction and Control
92. ICML 2025Trajectory World Models for Heterogeneous Environments
93. ICML 2025Video-Enhanced Offline Reinforcement Learning: A Model-Based Approach
94. ICML 2025What Has a Foundation Model Found? Inductive Bias Reveals World Models
95. ICML 2025When do neural networks learn world models?
96. ICML 2025World Model Implanting for Test-time Adaptation of Embodied Agents
97. ICML 20243D-VLA: A 3D Vision-Language-Action Generative World Model
98. ICML 2024AD3: Implicit Action is the Key for World Models to Distinguish the Diverse Visual Distractors
99. ICML 2024Do Transformer World Models Give Better Policy Gradients?
100. ICML 2024Efficient World Models with Context-Aware Tokenization

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。