OverSeer: Multi-Agent World Models Beyond What Agents See Paper • 2610.28554 • Published about 1 month ago • 7
OverSeer Collection Datasets of OverSeer: Multi-Agent World Models Beyond What Agents See (Minecraft and Doom, two players, fixed cameras). • 2 items • Updated Oct 12
Towards Physics of Multimodal Pretraining: Knowledge Flow, Modality Synergy, Early Unification, and Recipes Paper • 2608.05000 • Published Aug 6 • 60
Data Journalist Agent: Transforming Data into Verifiable Multimodal Stories Paper • 2606.11176 • Published Jun 9 • 131
Tuna-2: Pixel Embeddings Beat Vision Encoders for Multimodal Understanding and Generation Paper • 2604.24763 • Published Apr 27 • 70
WildWorld: A Large-Scale Dataset for Dynamic World Modeling with Actions and Explicit State toward Generative ARPG Paper • 2603.23497 • Published Mar 24 • 67
Grounding World Simulation Models in a Real-World Metropolis Paper • 2603.15583 • Published Mar 16 • 154