arXiv cs.AI· Xuhong Qiu, Zhuoyu Wu, Qiawei Zheng, Wenhui Ou, Wenqi Fang, Rapha\"{e}l C. -W. Phan·· 3 小时前AI 评分27
AGQ：面向视觉-语言-动作模型的动作引导超低比特量化
AGQ: Action-Guided Ultra-Low-Bit Quantization for Vision-Language-Action Models
AI 导读
AGQ 提出动作引导的超低比特量化方法，通过动作敏感度与权重恢复成本代理分配矩阵位宽，用于压缩视觉-语言-动作（VLA）模型。
正文 · 原文
Abstract:Vision-language-action (VLA) models enable robots to perform tasks specified through natural-language instructions, but their large weight footprint challenges deployment on memory-constrained onboard platforms. Ultra-low-bit quantization reduces this footprint, yet preserving task performance and controlling weight recovery costs remain challenging. We propose Action-Guided Quantization, which combines action sensitivity with a measured proxy for weight recovery cost to allocate matrix bitwidths under storage budgets. Per-channel calibration compensates for quantization error, while a uniform bitwidth within each matrix preserves regular layouts. Bit-dependent grouped lossless coding reduces resident weight storage, and block-wise recovery limits temporary buffers. On LIBERO, AGQ achieves 92.7% task success at W3, exceeding AWQ and GPTQ by 1.2 and 3.7 percentage points, respectively, and retains 70.5% success at W2.5. An asymmetric configuration, denoted W1.5$^{*}$, uses W4 vision weights and a parameter-weighted average of 1.5 bits for language weights. With lossless coding, it achieves $6.10\times$ complete-model compression relative to BF16 while retaining 25.5% task success. Code and experimental results are available at this https URL
| Comments: | 3 figures, 3 tables. Code and experimental results: this https URL |
| Subjects: | Computer Vision and Pattern Recognition (cs.CV); Artificial Intelligence (cs.AI) |
| Cite as: | arXiv:2610.33074 [cs.CV] |
| (or arXiv:2610.33074v1 [cs.CV] for this version) | |
| https://doi.org/10.48550/arXiv.2610.33074 arXiv-issued DOI via DataCite |
Submission history
From: Zhuoyu Wu [view email]
[v1]
Wed, 21 Oct 2026 06:22:51 UTC (908 KB)
来源：arXiv cs.AI · arxiv.org