On-the-fly Repulsion in the Contextual Space for Rich Diversity in Diffusion Transformers
Paper • 2603.28762 • Published • 22
Spring'26
-
-
Out of Sight but Not Out of Mind: Hybrid Memory for Dynamic Video World Models
Paper • 2603.25716 • Published • 147 -
ShotStream: Streaming Multi-Shot Video Generation for Interactive Storytelling
Paper • 2603.25746 • Published • 150 -
MACRO: Advancing Multi-Reference Image Generation with Structured Long-Context Data
Paper • 2603.25319 • Published • 32 -
AVControl: Efficient Framework for Training Audio-Visual Controls
Paper • 2603.24793 • Published • 25 -
Omni-WorldBench: Towards a Comprehensive Interaction-Centric Evaluation for World Models
Paper • 2603.22212 • Published • 124 -
Not All Layers Are Created Equal: Adaptive LoRA Ranks for Personalized Image Generation
Paper • 2603.21884 • Published • 5 -
WorldCache: Content-Aware Caching for Accelerated Video World Models
Paper • 2603.22286 • Published • 4 -
Versatile Editing of Video Content, Actions, and Dynamics without Training
Paper • 2603.17989 • Published • 16