CoPE-VideoLM: Codec Primitives For Efficient Video Language Models
Paper
•
2602.13191
•
Published
•
29
Video Perception
-
-
KTV: Keyframes and Key Tokens Selection for Efficient Training-Free Video LLMs
Paper • 2602.03615 • Published -
TiFRe: Text-guided Video Frame Reduction for Efficient Video Multi-modal Large Language Models
Paper • 2602.08861 • Published -
Causality-Aware Temporal Projection for Video Understanding in Video-LLMs
Paper • 2601.01804 • Published