FoldQuantVLA：通过一致性折叠实现视觉-语言-动作模型的原生低比特量化
FoldQuantVLA: Native Low-Bit Quantization of Vision-Language-Action Models via Consistent Folding
浏览论文内容
中文总结 AI 辅助
FoldQuantVLA通过一致性折叠实现视觉-语言-动作模型的训练后低比特量化，在保持机器人行为的同时，以W4A4配置在Orin和桌面GPU上获得1.2-1.52倍加速，并将真实任务成功率从80%提升至92.5%。
中文摘要 AI 辅助
低比特视觉-语言-动作推理必须在保持机器人行为的同时降低从观察到动作的延迟。我们提出了FoldQuantVLA，一个训练后量化框架，在标定、权重舍入和原生整数执行过程中保持一致的激活表示。它结合了通道缩放和块Hadamard变换与动态逐token量化，无需策略重训练。自定义TensorRT插件在语言主干和迭代动作专家中执行投影，在Ada GPU和Jetson AGX Orin上使用四比特权重和激活（W4A4）。评估涵盖LIBERO、SimplerEnv和两个机器人平台。在三个GR00T检查点和$\pi_{0.5}$上，W4A4在Orin上相比浮点TensorRT实现了$1.20$到$1.33\times$的加速，在桌面上实现了$1.25$到$1.52\times$的加速。将语言注意力输出和前馈下投影保留为八比特（W8A8）在所有四个检查点上提高了保留动作保真度。在四个真实机器人任务中，该配置将观察到的GR00T N1.7成功率从均匀W4A4的$80.0\\%$提高到每个配置80次试验中的$92.5\\%$，并测得Orin额外延迟为1毫秒。
英文摘要
Low-bit vision-language-action inference must reduce observation-to-action latency while preserving robot behavior. We present FoldQuantVLA, a post-training quantization framework that carries a consistent activation representation through calibration, weight rounding, and native integer execution. It combines channel scaling and block Hadamard transforms with dynamic per-token quantization, without policy retraining. Custom TensorRT plugins execute projections in both the language backbone and iterative action expert with four-bit weights and activations (W4A4) on Ada GPUs and Jetson AGX Orin. Evaluation spans LIBERO, SimplerEnv, and two robot platforms. Across three GR00T checkpoints and $π_{0.5}$, W4A4 achieves $1.20$ to $1.33\times$ speedups over floating-point TensorRT on Orin and $1.25$ to $1.52\times$ on desktop. Retaining language attention-output and feed-forward down projections at eight bits (W8A8) improves held-out action fidelity on all four checkpoints. Across four real-robot tasks, this configuration raises observed GR00T N1.7 success from $80.0\%$ with uniform W4A4 to $92.5\%$ over 80 trials per configuration, with a measured additional Orin latency of 1 ms.
发表机构
- VinRobotics
- University of Arkansas（阿肯色大学）
- Peking University（北京大学）
- VinUniversity
- TU Darmstadt（达姆施塔特工业大学）
机构由 AI 辅助整理，请以论文原文为准。
