AI Paper Atlas

Research map / Label-efficient and robust learning / Semi-supervised learning

From Local Geometry to Global Pseudo-Labeling for Robust Positive–Unlabeled Learning under Covariate Shift

Firas Gabetni, Alexandre Rocchi--Henry, Ziyi LIU, Nacim Belkhir, Gianni Franchi

ECCV 2026 · Label-efficient and robust learning › Semi-supervised learning · approach: Pseudo-labeling for specialized tasks · 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
add classifier-head distillation of kNN pseudo-labels (ablation)S-PUNA · ImageNet-V2AUROC: 92.65 -> 96.38Improved
add classifier-head distillation of kNN pseudo-labels (ablation)S-PUNA · ImageNet-V2FPR95: 33.70 -> 17.14Improved
add classifier-head distillation of kNN pseudo-labels (ablation)S-PUNA · EuroSAT-D (near shift)AUROC: 84.40 -> 99.79Improved
add classifier-head distillation of kNN pseudo-labels (ablation)S-PUNA · ImageNet-1K near shiftFPR95: 19.22 -> 9.69Improved
add classifier-head distillation of kNN pseudo-labels (ablation)S-PUNA · TinyImageNet near shiftAUROC: 100.00 -> 99.90No clear effect
use ViT block 6 features instead of block 11 (ablation)S-PUNA · ImageNet-1K averageAUROC: 97.03 -> 98.20Improved
increase shifted proportion of unlabeled pool from 5% to 50% (ablation)S-PUNA · ImageNet-1K averageAUROC: 91.61 -> 98.20Improved
increase shifted proportion from 50% to 75% (ablation)S-PUNA · ImageNet-1K averageAUROC: 98.20 -> 95.53Hurt
replace Dist-PU with S-PUNA at low shifted proportionPU baselines · ImageNet-1K average (p=5%)AUROC: 71.04 -> 91.61Improved
replace Dist-PU with S-PUNAPU baselines · ImageNet-1K near shiftFPR95: 35.05 -> 9.69Improved
replace Dist-PU with S-PUNAPU baselines · EuroSAT far shiftAUROC: 89.46 -> 100.00Improved
replace naive kNN propagation with symmetric S-PUNA expansion (no classifier)kNN baseline · ImageNet-1K near shiftAUROC: 54.40 -> 95.81Improved

Earlier methods this paper compares with

Named baselines from earlier accepted papers in this paper's results tables: better than 1 of 1 on most shared results.

Matched by method name in results tables read by AI. Authors' own tables tend to favour the newer method, and a comparison can be missed when a method is renamed.

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