EfficientVLA: Training-Free Acceleration and Compression for Vision-Language-Action Models

@misc{wang2025efficientvlatrainingfreeacceleration,
      title={EfficientVLA: Training-Free Acceleration and Compression for  Vision-Language-Action Models},
      author={Yuhao Wang and Linfeng Zhang and Zhipeng Zhang and Chuan Wen and Zichen Wen and Yantai Yang and Chang Zou and Luo Zhongwei},
      year={2025},
      eprint={2506.10100},
      archivePrefix={arXiv},
      primaryClass={cs.CV},
      url={https://arxiv.org/abs/2506.10100},
}