World Models
Overview
World models learn or encode how states, objects, and events evolve, often predicting what will happen under different actions or observations. These predictions can help agents plan ahead, learn with fewer real-world trials, test decisions safely, and reason about situations they cannot currently observe. Approaches range from latent-state predictors and action-conditioned video generators to object- or graph-based models, executable programs, and physics-informed dynamics. A central challenge is making predictions useful and reliable over long horizons: visual plausibility and one-step accuracy do not guarantee correct causal dynamics, persistent memory, or better decisions.
Latest papers
402 papers38 papers in the last four weeks, down 51% on the four weeks before. 0.3% of all new papers.
One Lens, Many Worlds : A Capability-Typed Interface for World-Model Interpretability
Unifying Object-Centric World Models and Diffusion Policy: A Hierarchical Framework for Multi-Stage Robotic Tasks
AnchorWorld: Embodied Egocentric World Simulation with View-based Evolution Customization
Learning Explicit Behavioral Models with Adaptive Questions and World-Model Probes
Towards World Models in Biomedical Research
Discrete-WAM: Unified Discrete Vision-Action Token Editing for World-Policy Learning
Flying on Imagination: DreamerV3-Style World Models under Environment Distribution Shift for Quadrotor Navigation
WAM-Nav: Asymmetric Latent World-Action Modeling for Unified Visual Navigation
MAD: Mapping-Aware World Models for Agile Quadrotor Flight
World Models Meet Language Models: On the Complementarity of Concrete and Abstract Reasoning
GeoSem-WAM: Geometry- and Semantic-Aware World Action Models
Exact equivariance, kept through training, buys zero-shot generalisation across the symmetry group
Cosmos 3: Omnimodal World Models for Physical AI
From Zero to Hero: Training-Free Custom Concept Spawning in World Models
Unified Driving Tokens: Representation- and Geometry-Guided Discrete Tokenizer for Driving World Models and Planning
RoboTrustBench: Benchmarking the Trustworthiness of Video World Models for Robotic Manipulation
-WM: A Unified Video-Action World Model for Robotic Manipulation
PatchWorld: Gradient-Free Optimization of Executable World Models
Physically Viable World Models: A Case for Query-Conditioned Embodied AI
Back to Parsimonious Latents: Learning Task-Centric World Models from Visual Foundations
LVDrive: Latent Visual Representation Enhanced Vision-Language-Action Autonomous Driving Model
stable-worldmodel: A Platform for Reproducible World Modeling Research and Evaluation
swm dramatically reduces research overhead and accelerates trustworthy progress toward reliable world models.