Why Can't I Open My Drawer? Mitigating Object-Driven Shortcuts in Zero-Shot Compositional Action Recognition Paper • 2601.16211 • Published 5 days ago • 3
Scenes as Objects, Not Primitives: Instance-Structured 3D Tokenization from Unposed Views Paper • 2606.29513 • Published 9 days ago • 50
Rethinking State Tracking in Recurrent Models Through Error Control Dynamics Paper • 2605.07755 • Published May 8 • 24
Decomposed Attention Fusion in MLLMs for Training-Free Video Reasoning Segmentation Paper • 2510.19592 • Published Oct 22, 2025 • 13
Multi-Granular Spatio-Temporal Token Merging for Training-Free Acceleration of Video LLMs Paper • 2507.07990 • Published Jul 10, 2025 • 45