提出混合精度量化框架，让视觉语言动作模型更省内存、更快推理。
Mix-QVLA: Task-Evidence-Aware Mixed-Precision Quantization of Vision-Language-Action Models
- 根据任务证据动态分配不同精度，关键层保留高精度
- 在LIBERO上内存减至4.1GB，推理速度提升1.52倍
- 适合部署低比特视觉语言动作模型的研究者与工程师
我们提出 Mix-QVLA，一种面向视觉语言动作（VLA）模型的任务证据感知混合精度量化方法。该方法将每个量化版本锚定在全精度动作标记参考决策上，评估量化是否在关键功能边界保持任务相关证据。通过边界激活计算归一化梯度加权任务证据图，使用证据质量与归因分布畸变比较全精度与量化结果，捕捉决策支持证据的强度和分配变化。引入软瓶颈目标函数，将边界级退化聚合为层级敏感度评分，并建模任务执行全过程中的敏感度动态变化，而非假设固定敏感性。最终的证据与时序感知评分指导在模型尺寸与位操作预算下的混合精度比特分配。在类 OpenVLA 策略上广泛验证显示，该方法显著提升低比特部署的准确率-效率平衡。在 LIBERO 数据集上，混合精度方案将 OpenVLA-OFT 内存从 15.4 GB 降至 4.1 GB，平均成功率保持 96.3%（全精度为 97.1%），推理速度提升 1.52 倍。
原文摘要 · Abstract (English)
We propose Mix-QVLA, a task-evidence-aware mixed-precision PTQ framework for VLA models. Mix-QVLA anchors each quantized variant to the full-precision action-token reference decision and evaluates whether quantization preserves task-relevant evidence across key VLA functional boundaries. It computes normalized gradient-weighted task-evidence maps from boundary activations and compares full-precision and quantized maps using evidence-mass and attribution-distribution distortion, capturing changes in both the strength and allocation of decision-supporting evidence. A soft-bottleneck objective aggregates boundary-level degradation into layer-wise sensitivity scores. Mix-QVLA further models sensitivity throughout task execution, capturing phase-dependent shifts in layer importance rather than assuming a fixed sensitivity profile. The resulting evidence- and time-aware scores guide mixed-precision bit allocation under model-size and BitOps budgets. Extensive evaluations on OpenVLA-style policies show that Mix-QVLA improves the accuracy-efficiency trade-off of low-bit VLA deployment. On LIBERO, Mix-QVLA reduces OpenVLA-OFT memory from 15.4 GB to 4.1 GB, retains 96.3 average success compared with 97.1 for the BF16 model, and achieves a 1.52x inference speedup.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。