AI Paper Atlas

Research map / Image generation / Fast diffusion sampling

Q&C: When Quantization Meets Cache in Efficient Generation

Xin Ding, Xin Li, Haotong Qin, Zhibo Chen

ICLR 2026 · Image generation › Fast diffusion sampling · approach: Distillation for few-step generation

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
naively stack quantization (PTQ4DiT) and cache (Learn-to-Cache) (ablation)PTQ4DiT · ImageNet 256 W8A8 50 stepsFID: 5.45 -> 13.67Hurt
add Temporal-Aware Parallel Clustering for calibration data (ablation)Quantization+cache baseline · ImageNet 256 W8A8 50 stepsFID: 13.67 -> 8.34Improved
add Variance Compensation (ablation)Quantization+cache baseline · ImageNet 256 W8A8 50 stepsFID: 13.67 -> 9.65Improved
combine TAP and VC (ablation)Quantization+cache baseline · ImageNet 256 W8A8 50 stepsFID: 13.67 -> 5.43Improved
use TAP instead of K-Means for calibration selection (ablation)Q&C · ImageNet 256 W8A8 50 stepsFID: 10.31 -> 8.34Improved
use TAP instead of Agglomerative clustering (ablation)Q&C · ImageNet 256 W8A8 50 stepsFID: 9.56 -> 8.34Improved
vary spatial/temporal weight alpha in TAP from 0.3 to 0.7 (ablation)TAP · ImageNet 256 W8A8 50 stepsFID: 5.57 / 5.46 / 5.43 / 5.36 / 5.45No clear effect
combine quantization and cache (Q&C) vs PTQ4DiT alone at 50 stepsPTQ4DiT · ImageNet 256 W8A8 50 stepsspeedup: 10x -> 12.7xImproved
apply Q&C to 512x512 W4A8PTQ4DiT · ImageNet 512 W4A8 100 stepsFID: 19.00 -> 19.05; speed 2.5x -> 4xNo clear effect
apply Q&C to latent diffusion on LSUN-ChurchAPQ-DM · LSUN-Church W8A8FID: 9.04 -> 9.10No clear effect
combined quantization and cache raises calibration sample similarityQuantization+cache · DiT calibration samplescosine similarityHurt

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