world models
A representation of an environment used by an AI agent to simulate its interactions, often enabling it to plan actions and make decisions based on imagined experiences.
- COME: Adding Scene-Centric Forecasting Control to Occupancy World Model
- DMWM: Dual-Mind World Model with Long-Term Imagination
- Deep RL Needs Deep Behavior Analysis: Exploring Implicit Planning by Model-Free Agents in Open-Ended Environments
- DyMoDreamer: World Modeling with Dynamic Modulation
- Dyn-O: Building Structured World Models with Object-Centric Representations
- EDELINE: Enhancing Memory in Diffusion-based World Models via Linear-Time Sequence Modeling
- Force Prompting: Video Generation Models Can Learn And Generalize Physics-based Control Signals
- From Forecasting to Planning: Policy World Model for Collaborative State-Action Prediction
- Imagined Autocurricula
- Learning 3D Persistent Embodied World Models
- Overcoming Challenges of Long-Horizon Prediction in Driving World Models
- PoE-World: Compositional World Modeling with Products of Programmatic Experts
- RLVR-World: Training World Models with Reinforcement Learning
- Revisiting Multi-Agent World Modeling from a Diffusion-Inspired Perspective
- RoboScape: Physics-informed Embodied World Model
- SAMPO: Scale-wise Autoregression with Motion Prompt for Generative World Models
- SPARTAN: A Sparse Transformer World Model Attending to What Matters
- Social World Model-Augmented Mechanism Design Policy Learning
- StateSpaceDiffuser: Bringing Long Context to Diffusion World Models
- Towards foundational LiDAR world models with efficient latent flow matching
- Video World Models with Long-term Spatial Memory
- ViewPoint: Panoramic Video Generation with Pretrained Diffusion Models
- WALL-E: World Alignment by NeuroSymbolic Learning improves World Model-based LLM Agents
- World Models Should Prioritize the Unification of Physical and Social Dynamics
- WorldModelBench: Judging Video Generation Models As World Models
- Zero-shot World Models via Search in Memory