WanderBench: Can World Models Explore Dynamic Worlds from Video?
cs.CV, cs.AI
Submitted: 2026-11-09
Updated: 2026-11-09
Terminology
Sources
- MMGR: Multi-Modal Generative Reasoning Benchmark and Evaluation
- The Matrix: Infinite-Horizon World Generation with Real-Time Moving Control
- "PhyWorldBench": A Comprehensive Evaluation of Physical Realism in Text-to-Video Models
- Matrix-game 2.0: An open-source, real-time, and streaming interactive world model
- GAIA-1: A Generative World Model for Autonomous Driving
- Vid2World: Crafting Video Diffusion Models to Interactive World Models
- 3D and 4D World Modeling: A Survey
- WorldModelBench: Judging Video Generation Models As World Models
- Hunyuan-GameCraft: High-dynamic Interactive Game Video Generation with Hybrid History Condition
- CameraAnything: Refilming Videos with Arbitrary Camera Control
- Sekai: A Video Dataset towards World Exploration
- WorldLens: Full-Spectrum Evaluations of Driving World Models in Real World
- Depth Anything 3: Recovering the Visual Space from Any Views
- 4DWorldBench: A Comprehensive Evaluation Framework for 3D/4D World Generation Models
- HPSv3: Towards Wide-Spectrum Human Preference Score
- Yume: An Interactive World Generation Model
- WorldSimBench: Towards Video Generation Models as World Simulators
- Advancing Open-source World Models
- WorldArena: A Unified Benchmark for Evaluating Perception and Functional Utility of Embodied World Models
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