Safety
VisionMX: Unlocking Microscaling Post-Training Quantization for Vision Models
arXiv:2610.03218v1 Announce Type: new Abstract: Microscaling (MX) formats are emerging as a hardware-supported approach to efficient training and inference. They combine low-precision elements with sh
arXiv:2610.03218v1 Announce Type: new Abstract: Microscaling (MX) formats are emerging as a hardware-supported approach to efficient training and inference. They combine low-precision elements with shared block scales, but their impact on vision models remains underexplored. We systematically investigate post-training MX quantization across vision models and tasks. An analysis of direct conversion identifies three sources of error: block-scale representation, the poor alignment of some small convolutional weight tensors with nonuniform element grids, and the underuse of signed codes by nonnegative activations. These findings motivate VisionMX, a post-training MX quantization method that optimizes bounded weight rounding and applies a foldable affine correction to activations. We evaluate VisionMX across image classification, object detection, semantic segmentation, and low-light image enhancement using several MX-style formats. It improves on direct conversion and the evaluated post-training quantization baselines, with the largest performance recoveries in architectures most sensitive to MX conversion
Related
- Beyond Reconstruction Loss in Post-Training Quantization: Balanced Fitting for Large Vision-Language Models
- CHASE-VLA: Post-Training Quantization Framework for Vision-Language-Action Models with Chunk-Aware Scale Estimation
- RAMP: Robust Adaptive Mixed-Precision Quantization for Edge CPU Vision Models
- Efficient Quantization-Aware Distillation with Cross-Modal Alignment for Edge Vision-Language Models
Source: arXiv cs.CV | 2026-10-05