AI Paper Atlas

Research map / Probabilistic and scientific modeling / Uncertainty and Gaussian processes

Evidential Transformation Network: Turning Pretrained Models into Evidential Models for Post-hoc Uncertainty Estimation

Yongchan Chun, Chanhee Park, Jeongho Yoon, Jaehyung Seo, Heuiseok Lim

CVPR 2026 · Probabilistic and scientific modeling › Uncertainty and Gaussian processes · approach: Epistemic versus aleatoric decomposition · 1 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 ETN instead of fine-tuning with CE loss on ImageNetResNet50 (ImageNet) · ImageNetaccuracy: 46.03 -> 79.61Improved
use ETN instead of MAP_EDL fine-tuning on CIFAR-10 OOD (SVHN)VGG16 · CIFAR-10 -> SVHNAUPR (MI): 76.42 -> 85.22Improved
use ETN instead of Laplace approximation on ImageNet-A OODResNet50 · ImageNet -> ImageNet-AAUPR (MI): 96.36 -> 93.98Hurt
use sample-dependent ETN transformation instead of static scaling or AdaTS on ImageNetETN · ImageNetUncertainty Mass AUPR: about 17% and 10% drops in UM for static scaling and AdaTSImproved
use ETN instead of static scaling or AdaTS for OOD mutual information on ImageNetETN · ImageNetAUPR (MI) OOD: 21.5% and 25.5% lower MI for static scaling and AdaTSImproved
use ETN instead of static scaling or AdaTS for mutual information on OBQA (ablation)ETN · OBQAAUPR (MI) OOD: 13.6% and 24.7% lower MI for static scaling and AdaTSImproved
use ETN instead of static scaling on CIFAR-10 mutual information (ablation)ETN · CIFAR-10AUPR (MI) OOD: static scaling trails ETN by roughly 5%Improved
make the transformation sample-dependent rather than static (ablation)ETN / AdaTS vs static scaling · ImageNetuncertainty estimationImproved
use scalar instead of vector or matrix transformation parameter (ablation)ETN · OBQAID and OOD AUPRImproved
use matrix instead of scalar transformation parameter (ablation)ETN · CIFAR-10OOD AUPRMixed
use vector instead of scalar parameter (ablation)ETN · RACE vs ImageNetAUPRMixed
increase the mode of the prior p(A) (ablation)ETN · CIFAR-10 and RACEAUPRImproved

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