AI Paper Atlas

Research map / Architectures and efficiency / Quantization

Distance-Aware Quantization

Dohyung Kim, Junghyup Lee, Bumsub Ham

ICCV 2021 · Architectures and efficiency › Quantization · approach: Mixed-precision quantization · 38 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
replace fixed temperature beta=4 with adaptive temperature beta* (ablation)DASR (ResNet-20) · CIFAR-10 test (W/A 1/1)top-1 accuracy (%): 13.5 -> 85.8Improved
replace fixed temperature beta=12 with adaptive temperature (ablation)DASR (ResNet-20) · CIFAR-10 test (W/A 1/1)top-1 accuracy (%): 69.9 -> 85.8Improved
use DASR at test time with beta=4 instead of rounding (ablation)DASR (ResNet-20) · CIFAR-10 test (W/A 1/1)top-1 accuracy (%): 13.5 -> 89.7Improved
raise fixed temperature from 12 to 24 (ablation)DASR (ResNet-20) · CIFAR-10 test (W/A 1/1)top-1 accuracy (%): 69.9 -> 59.6Hurt
use kernel soft argmax instead of plain soft argmax at beta=10 vs 12 (ablation)DASR (ResNet-20) · CIFAR-10 test (W/A 1/1)top-1 accuracy (%): 11.7 -> 69.9Improved
replace temperature annealing with temperature controller (ablation)DAQ (ResNet-20) · CIFAR-10 test (W/A 1/1)top-1 accuracy (%): 72.2 -> 85.8Improved
replace combined STE with temperature controller (ablation)DAQ (ResNet-20) · CIFAR-10 test (W/A 1/1)top-1 accuracy (%): 81.7 -> 85.8Improved
DAQ vs best prior at 1/1 ResNet-18 (QNet)DAQ · ImageNet val ResNet-18 (W/A 1/1)top-1 accuracy (%): 53.6 -> 56.2Improved
DAQ vs IRNet at 1/32 ResNet-18DAQ · ImageNet val ResNet-18 (W/A 1/32)top-1 accuracy (%): 66.5 -> 67.2Improved
DAQ vs LSQ+ at 4/4 ResNet-18DAQ · ImageNet val ResNet-18 (W/A 4/4)top-1 accuracy (%): 70.8 -> 70.5No clear effect
DAQ vs PROFIT on MobileNet-V2 4/4DAQ · ImageNet val MobileNet-V2 (W/A 4/4)top-1 accuracy (%): 71.6 -> 70.0Hurt
DAQ vs IRNet at 1/1 ResNet-20DAQ · CIFAR-10 test ResNet-20 (W/A 1/1)top-1 accuracy (%): 85.4 -> 85.8Improved

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