PlayGen2: Learning Real-Time Playable Worlds with Explicit 3D State
cs.CV, cs.LG
Submitted: 2026-11-05
Updated: 2026-11-05
Code: https://github.com/MajorDavidZhang/PlayGen-2
Project page: https://nvlabs.github.io/Sana/Sol-Engine
Terminology
Sources
- MASS: Multiplayer World Models with Authoritative Shared State
- Code World Model: Coding Agent as World Brain
- AgentGarten: Code Worlds for Evolving Agents
- LiveWorld: Simulating Out-of-Sight Dynamics in Generative Video World Models
- Code Plans, Diffusion Renders: Open-Ended Generative World Modeling
- Infinite Worlds with Versatile Interactions
- Matrix-game 2.0: An open-source, real-time, and streaming interactive world model
- Programmable World Model
- ViZDoom: A Doom-based AI Research Platform for Visual Reinforcement Learning
- V-Net: Fully Convolutional Neural Networks for Volumetric Medical Image Segmentation
- MultiGen: Level-Design for Editable Multiplayer Worlds in Diffusion Game Engines
- Advancing Open-source World Models
- Diffusion Models Are Real-Time Game Engines
- WorldProbe: When Video World Models Fall Apart under Long-Horizon Spatial Exploration
- Matrix-Game 3.0: Real-Time and Streaming Interactive World Model with Long-Horizon Memory
- Playable Game Generation
- Matrix-Game 4.0: Harnessing Video Models for Real-Time Playable Game Worlds
- Matrix-Game: Interactive World Foundation Model
Related papers
- TEA: Text Encoder Alignment for Robust Concept Erasure in Text-to-Image Models
- Do Egocentric Video-Language Models Capture Both Hand- and Object-Centric Cues?
- Aura: Consistent Multi-Subject Video Generation via VLM-Grounded Semantic Alignment
- PVRF: All-in-one Adverse Weather Removal via Prior-modulated and Velocity-constrained Rectified Flow
- DLWM: Diverse Latent World Models for Efficient Multimodal Reasoning
- Out of Sight, Still in Mind: Token Compression for Omni-LLMs