"visual token selection" Papers
2 papers found
Conference
EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Models
Yantai Yang, Yuhao Wang, Zichen Wen et al.
NEURIPS 2025oralarXiv:2506.10100
34
citations
Principles of Visual Tokens for Efficient Video Understanding
Xinyue Hao, Li, Shreyank Gowda et al.
ICCV 2025arXiv:2411.13626
1
citations