EaqVLA: Encoding-aligned Quantization for Vision-Language-Action Models
Fuente:
arXiv
Gespeichert in:
| Hauptverfasser: | , , , , , |
|---|---|
| Format: | Preprint |
| Veröffentlicht: |
2025
|
| Schlagworte: | |
| Online-Zugang: | |
| Tags: |
Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
|
| _version_ | 1866909712593715200 |
|---|---|
| author | Jiang, Feng Zheng, Zihao Cui, Xiuping Li, Maoliang Chen, JIayu Chen, Xiang |
| author_facet | Jiang, Feng Zheng, Zihao Cui, Xiuping Li, Maoliang Chen, JIayu Chen, Xiang |
| contents | With the development of Embodied Artificial intelligence, the end-to-end control policy such as Vision-Language-Action (VLA) model has become the mainstream. Existing VLA models faces expensive computing/storage cost, which need to be optimized. Quantization is considered as the most effective method which can not only reduce the memory cost but also achieve computation acceleration. However, we find the token alignment of VLA models hinders the application of existing quantization methods. To address this, we proposed an optimized framework called EaqVLA, which apply encoding-aligned quantization to VLA models. Specifically, we propose an complete analysis method to find the misalignment in various granularity. Based on the analysis results, we propose a mixed precision quantization with the awareness of encoding alignment. Experiments shows that the porposed EaqVLA achieves better quantization performance (with the minimal quantization loss for end-to-end action control and xxx times acceleration) than existing quantization methods. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2505_21567 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | EaqVLA: Encoding-aligned Quantization for Vision-Language-Action Models Jiang, Feng Zheng, Zihao Cui, Xiuping Li, Maoliang Chen, JIayu Chen, Xiang Computer Vision and Pattern Recognition Machine Learning With the development of Embodied Artificial intelligence, the end-to-end control policy such as Vision-Language-Action (VLA) model has become the mainstream. Existing VLA models faces expensive computing/storage cost, which need to be optimized. Quantization is considered as the most effective method which can not only reduce the memory cost but also achieve computation acceleration. However, we find the token alignment of VLA models hinders the application of existing quantization methods. To address this, we proposed an optimized framework called EaqVLA, which apply encoding-aligned quantization to VLA models. Specifically, we propose an complete analysis method to find the misalignment in various granularity. Based on the analysis results, we propose a mixed precision quantization with the awareness of encoding alignment. Experiments shows that the porposed EaqVLA achieves better quantization performance (with the minimal quantization loss for end-to-end action control and xxx times acceleration) than existing quantization methods. |
| title | EaqVLA: Encoding-aligned Quantization for Vision-Language-Action Models |
| topic | Computer Vision and Pattern Recognition Machine Learning |
| url | https://arxiv.org/abs/2505.21567 |