Preprint
Sep 2026
SubRot: Signed Gradient Subspace Calibration for VLM Rotation Quantization
Post-training quantization reduces the deployment cost of vision-language models (VLMs), but preserving multimodal capabilities at low bit widths remains challenging. Existing methods rely on modality- or token-level gradient statistics, which are susceptible to cross-sample variations in visual-to-textual token ratios...
Zhen-Hao Shang, Hai-Zhao Jing, Hao-Kui Zhang et al.
· 0 citations