Imaginative Perception Tokens Enhance Spatial Reasoning in Multimodal Language Models
Paper • 2606.03988 • Published • 115
2606
-
-
Cosmos 3: Omnimodal World Models for Physical AI
Paper • 2606.02800 • Published • 119 -
Agents' Last Exam
Paper • 2606.05405 • Published • 324 -
ABot-Earth 0.5: Generative 3D Earth Model
Paper • 2606.09967 • Published • 295 -
Latent Spatial Memory for Video World Models
Paper • 2606.09828 • Published • 62