SOURCE-LINKED INTELLIGENCE

MiX: Micro-Inverted-Scaling for End-to-End Low-Bit Vision-Language Model Acceleration

arXiv · AI, language, vision and robotics · article · Sep 17, 2026 · UTC

The deployment of Vision-Language Models (VLMs) on edge devices is severely bottlenecked by memory bandwidth, necessitating aggressive sub-8-bit quantization. Since edge accelerators are strictly constrained by area and power, they require end-to-end quantized models. However, the extreme dynamic range gap between multi-modal tokens causes standard block formats to suffer "microscaling collapse," where a single massive outlier hijacks the shared exponent, underflowing surrounding elements and destroying attention maps. To break this bottleneck, we propose Micro-Inverted-Scaling (MiX), a novel

Read original source ↗ Open in workspace

recordType

paper

region

Global