AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

HQ-DM: Single Hadamard Transformation-Based Quantization-Aware Training for Low-Bit Diffusion Models

Shizhuo Mao, Hongtao Zou, Qihu Xie, Song Chen, Yi Kang

ECCV 2026 · Image generation › Fast diffusion sampling · approach: Training-free diffusion acceleration · 0 citations (OpenAlex)

Read the paperCheck your idea against 137,217 papers

AI-extracted. The measured effects below were read by AI from the paper's full text (no summary yet). Numbers and claims can be wrong or missing context, so check the paper before you cite or build on it.

Measured effects reported in the paper

What was changedWhereMeasuredEffect
use Single instead of Double Hadamard transformation (ablation)HQ-DM W4A4 · ImageNet 256x256 LDM-4IS: 4.37 -> 248.30Improved
use Single instead of Double Hadamard transformation (ablation)HQ-DM W4A4 · ImageNet 256x256 LDM-4FID: 179.40 -> 6.58Improved
use Single instead of Double Hadamard transformation (ablation)HQ-DM W8A8 · ImageNet 256x256 LDM-4latency: 1.0688 -> 0.9419 s/sample (1.13x)Improved
add Hadamard transform to naive PTQ + LSQ (ablation)Naive PTQ · ImageNet 256x256 LDM-4 W4A4IS: 5.09 -> 33.29Improved
add Hadamard transform to naive PTQ + LSQ (ablation)Naive PTQ · ImageNet 256x256 LDM-4 W4A4FID: lowering the FID by 125.43Improved
add Hadamard transform to naive QAT + LSQ (ablation)Naive QAT · ImageNet 256x256 LDM-4 W4A4IS: 159.24 -> 203.69Improved
add Hadamard transform to naive QAT + LSQ (ablation)Naive QAT · ImageNet 256x256 LDM-4 W4A4sFID: 8.38 -> 10.98Hurt
add Hadamard transform to LoRA only (ablation)LoRA distillation · ImageNet 256x256 LDM-4 W4A4FID: 8.55 -> 6.52Improved
add Hadamard transform to LoRA only (ablation)LoRA distillation · ImageNet 256x256 LDM-4 W4A4IS: 185.51 -> 229.75Improved
add learnable LSQ scales to LoRA + Hadamard (ablation)HQ-DM · ImageNet 256x256 LDM-4 W4A4IS: 229.75 -> 248.30Improved
lower activation bits from W4A4 to W4A3HQ-DM vs EfficientDM · ImageNet 256x256 LDM-4IS: 220.20 -> 248.30 at W4A4; 8.74 -> 49.62 at W4A3Improved
HQ-DM vs EfficientDM on LSUN-ChurchesHQ-DM · LSUN-Churches LDM-8FID: 12.83 -> 9.64 (W4A4); 20.97 -> 14.67 (W4A3)Improved

Closest papers by meaning

Writing a paper in this area? scime draft is a free LaTeX editor with AI included, real-time co-authors and a check of your references and venue rules before you submit. It also imports OpenAI Prism projects.

Open scime draft

Get the next scime Dispatch and a monthly field update

Free. We send a confirmation link first; every email has a one-click unsubscribe. Privacy