AI Paper Atlas

Research map / Label-efficient and robust learning / Recognition and long-tail

Quantile‑Adaptive Temperature Scaling for Confidence Calibration

Omprakash Chakraborty, Leo Fillioux, Ismail Ayed, Jose Dolz

ECCV 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Confidence calibration objectives · 2 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 piecewise linear temperature (K=10) instead of linear (ablation)QaTS · CIFAR-100-LT R-100 (ViT-B)ECE (%): 8.37 -> 8.76Hurt
use piecewise linear temperature (K=10) instead of linear (ablation)QaTS · CIFAR-100-C (ViT-B)ECE (%): 9.89 -> 10.64Hurt
use piecewise linear temperature (K=2) instead of linear (ablation)QaTS · CIFAR-100-C (ViT-B)ECE (%): 9.89 -> 10.14Hurt
use piecewise linear temperature (K=4) instead of linear (ablation)QaTS · CIFAR-100 clean (ViT-B)ECE (%): 2.26 -> 2.24No clear effect
use piecewise linear temperature (K=4) instead of linear (ablation)QaTS · CIFAR-100-LT R-100 (ViT-B)ECE (%): 8.37 -> 8.68Hurt
apply QaTS on top of Brier-loss trainingBrier-trained RN-50 · CIFAR-100ECE (%): 5.54 -> 2.72Improved
replace FeatClip with QaTS (accuracy preservation)FeatClip · 161 dataset-architecture combinationsaccuracy: QaTS higher accuracy in 94% of settingsImproved
replace FeatClip with QaTS (segmentation)FeatClip · VOC-2012mIoU: 69.78 -> 69.51Improved
replace FeatClip with QaTS (text)FeatClip · NewsGroup-20accuracy: 84.9 -> 84.7Improved
use QaTS instead of FeatClip under mild long-tailFeatClip · CIFAR-100-LT R-10, RN-50ECE (%): 6.62 -> 3.76Improved
use QaTS instead of TSTS · CIFAR-100 (RN-50)ECE (%): 5.32 -> 2.56Improved
use QaTS instead of FeatClip on ViT-S CIFAR-10FeatClip · CIFAR-10 (ViT-S)ECE (%): 0.62 -> 0.51Improved

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