AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Generalizable Mixed-Precision Quantization via Attribution Rank Preservation

Ziwei Wang, Han Xiao, Jiwen Lu, Jie Zhou

ICCV 2021 · Architectures and efficiency › Quantization · approach: Ultra-low-bit LLM quantization · 27 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
search policy with attribution rank preservation on CIFAR-10 instead of HAQ/EdMIPS search on CIFAR-10mixed-precision quantization · ImageNet MobileNet-V2 (~7-8 GBOPs)top-1 accuracy: 62.7 -> 70.4Improved
add GMPQ rank-preservation to policy found on CIFAR-10EdMIPS-C · ImageNet ResNet50 (~16 GBOPs)top-1 accuracy: 65.6 -> 73.6Improved
add GMPQ rank-preservation to policy found on CIFAR-10EdMIPS-C · ImageNet ResNet18 (~7 GBOPs)top-1 accuracy: 59.1 -> 67.8Improved
search on CIFAR-10 with GMPQ instead of on ImageNet with EdMIPSResNet50 · ImageNet (~16 GBOPs)search cost (GPU hours): 26.5 -> 3.4Improved
search on CIFAR-10 with GMPQ instead of on ImageNet with HAQMobileNet-V2 · ImageNet (~7-8 GBOPs)search cost (GPU hours): 51.1 -> 2.6Improved
use GMPQ instead of HAWQResNet18 · ImageNet (~28-34 GBOPs)top-1 accuracy: 68.5 -> 70.2Improved
use GMPQ with policy searched on CIFAR-10 vs HAQ-CSSD VGG16 · PASCAL VOC07 (~420 GBOPs)mAP: 66.4 -> 69.2Improved
use GMPQ with policy searched on CIFAR-10 vs HAQ-CFaster R-CNN ResNet18 · PASCAL VOC07 (~290 GBOPs)mAP: 70.4 -> 73.4Improved
use GMPQ instead of HAQ searched on VOCSSD VGG16 · PASCAL VOC07 (~800 GBOPs)mAP: 70.9 -> 70.5Hurt
use capacity-aware p instead of fixed p in attribution imitation (ablation)GMPQ ResNet18 · ImageNet accuracy-complexity trade-offaccuracy-complexity curveImproved
use medium fixed p (versus small or large p) (ablation)GMPQ · ImageNet ResNet18accuracy-complexity curveImproved
use medium generalization-risk weight eta (ablation)GMPQ · ImageNet ResNet18accuracy-complexity curveImproved

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