3 papers
cs.CV2026
FlashGaze: Training-Free Multi-Scale Patch Pruning For Efficient Video Understanding
Ziye Zhu, Yanghao Zhou, Lixing Tan +3
Multimodal Large Language Models (MLLMs) have demonstrated strong performance in video understanding, yet efficiently processing long, high-resolution videos remains challenging. S…
cs.CV2026
Can Motion-Language Models Ground Structure? STRIDE for Evaluating the Evaluators
Lixing Tan, Qing Xia, Yuting Guo +2
Motion-language models are typically scored by motion-language evaluators, but how well these evaluators ground language structure remains unclear. Here, we introduce the Structure…
cs.CV2026
REMEDY: How Far Is Video Generation from Medical Education World Models?
Lixing Tan, Yanghao Zhou, Qing Xia +3
Recent video generation models produce realistic videos and show potential as a foundation for world models. These advances create opportunities for generating medical teaching dem…