AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Differentiable Vector Quantization for Rate-Distortion Optimization of Generative Image Compression

Shiyin Jiang, Wei Long, Minghao Han, Zhenghao Chen, Ce Zhu, Shuhang Gu

CVPR 2026 · Architectures and efficiency › Quantization · approach: Post-training quantization of ViTs · 3 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
add differentiable relaxation of the index distribution for the rate loss (ablation)RDVQ · DIV2K-valDISTS: 0.2147 -> 0.1005Improved
add differentiable relaxation of the index distribution for the rate loss (ablation)RDVQ · DIV2K-valFID: 86.93 -> 19.96Improved
add differentiable relaxation of the index distribution for the rate loss (ablation)RDVQ · DIV2K-valLPIPS: 0.5031 -> 0.2321Improved
add differentiable relaxation of the index distribution for the rate loss (ablation)RDVQ · DIV2K-valbpp: 0.0464 -> 0.0247Improved
use joint RD learning instead of K-means codebook-size rate control (ablation)RDVQ · DIV2K-valDISTS: 0.1253 -> 0.1005Improved
use joint RD learning instead of K-means codebook-size rate control (ablation)RDVQ · DIV2K-valFID: 28.08 -> 19.96Improved
use joint RD learning instead of K-means codebook-size rate control (ablation)RDVQ · DIV2K-valLPIPS: 0.2831 -> 0.2321Improved
jointly optimize autoencoder and entropy model (vs separate AE-Entropy) for prefix-based rate adjustment (ablation)RDVQ-Adj · Kodak (test-time rate adjustment)R-D curve / DISTSImproved
use autoregressive completion instead of zero-padding for missing features (ablation)RDVQ-Adj · Kodak (test-time rate adjustment)R-D curve / DISTSImproved
use piecewise tau/lambda (tau=0.1 at ultra-low bpp) instead of fixed sharp tau=0.01 (ablation)RDVQ · DIV2K-valRD performanceImproved
add high-resolution fine-tuning on OpenImage then DF2K after ImageNet (ablation)RDVQ · DIV2K-valreconstruction qualityImproved

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