AI Paper Atlas

Research map / Label-efficient and robust learning / Domain adaptation and generalization

PAS: Estimating the target accuracy before domain adaptation

Raphaella Diniz, Jackson de Faria, Martin Ester

ICLR 2026 · Label-efficient and robust learning › Domain adaptation and generalization · approach: Source-free domain adaptation · 0 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 PAS instead of MMD to predict target accuracysource selection for UDA · All benchmarksPearson correlation: 0.37 -> 0.83Improved
use PAS instead of A-distancesource selection for UDA · All benchmarksPearson correlation: 0.04 -> 0.83Improved
use cosine distance to class centroid instead of Euclidean distance (ablation)PAS · All benchmarksPearson correlation: 0.68 -> 0.79Improved
use cosine distance to class centroid instead of Euclidean distance (ablation)PAS · Office-31Pearson correlation: 0.69 vs 0.63Hurt
use class centroid instead of average cosine distance to source samples (ablation)PAS · All benchmarksPearson correlation: 0.66 -> 0.79Improved
apply PAS to multi-object imagesPAS · ImageCLEFPearson correlation: 0.44 overall; 0.22 for ResNet-50Hurt
use subset of samples to compute PAS (ablation)PAS · Office-HomePAS value / rankingNo clear effect
use PAS to rank pre-trained backbones with fixed DA methodPAS with DANN and MCC · Office-Home A->C, Office-31 W->Atarget accuracy rankingImproved
use PAS instead of MMD on DomainNetsource selection for UDA · DomainNetPearson correlation: -0.09 -> 0.53Improved

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