From Game Generation to Game Playing: Fully Automated Ability Training for Vision-Language-Action Models
cs.CV
Submitted: 2026-11-08
Updated: 2026-11-19
Terminology
Sources
- Gym-Anything for Robotics: Scaling physical simulations with digital agents
- Gym-Anything-RT: Scaling Real-Time Environments for Vision-Language Agents
- MineStudio: A Streamlined Package for Minecraft AI Agent Development
- GameDevBench: Evaluating Agentic Capabilities Through Game Development
- GameWAM: A World Action Model for Video Games
- Isekai3D SDG: A Platform for Diverse Synthetic Gameplay Visual Data
- OmniGameArena: A Unified UE5 Benchmark for VLM Game Agents with Improvement Dynamics
- GameCraft-Bench: Can Agents Build Playable Games End-to-End in a Real Game Engine?
- NitroGen: An Open Foundation Model for Generalist Gaming Agents
- GameWorld: Towards Standardized and Verifiable Evaluation of Multimodal Game Agents
- Player-1: Learning to Play from Synthetic Experience
- UI-TARS: Pioneering Automated GUI Interaction with Native Agents
- Scaling Instructable Agents Across Many Simulated Worlds
- SIMA 2: A Generalist Embodied Agent for Virtual Worlds
- Cradle: Empowering Foundation Agents Towards General Computer Control
- Lumine: An Open Recipe for Building Generalist Agents in 3D Open Worlds
- HiSTEVE 1.0: Harnessing Human-Interface VLA Agents for Open-World Games
- TaskFrame: Connecting Task-Level Planning with Frame-Level Control for Game Agents
- OpenHA: A Series of Open-Source Hierarchical Agentic Models in Minecraft
- Game-TARS: Pretrained Foundation Models for Scalable Generalist Multimodal Game Agents
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