AI Paper Atlas

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

OPAL: Orthonormal Prototype Alignment Learning for Interpretable Image Classification

Ilán Carretero, Gustavo ANGULO, Rocío Amor, Valery Naranjo

ECCV 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Prototype and concept dictionary models · 0 citations (OpenAlex)

Read the paper

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 channel-wise softmax (ablation)OPAL (Proj+GMP+Orth) · CUB spatial prototype collapseSPC: 77.3 -> 1.7Improved
add channel-wise softmax (ablation)OPAL (Proj+GMP+Orth) · DOGS spatial prototype collapseSPC: 93.6 -> 4.9Improved
add channel-wise softmax (ablation)OPAL (Proj+GMP+Orth) · PETStop-1 accuracy (%): 93.2 -> 92.8Hurt
add channel-wise softmax (ablation)OPAL (Proj+GMP+Orth) · CUBtop-1 accuracy (%): 82.0 -> 85.6Improved
add orthonormal anchors (ablation)Proj-only network · CUBtop-1 accuracy (%): 82.3 -> 85.0Improved
add orthonormal anchors (ablation)Proj-only network · CARStop-1 accuracy (%): 86.3 -> 90.6Improved
add global max pooling without orthonormal anchors (ablation)ConvNeXt-Tiny · CARStop-1 accuracy (%): 87.8 -> 75.7Hurt
add global max pooling to Proj+CWS+Orth (ablation)OPAL · CARStop-1 accuracy (%): 90.4 -> 90.3No clear effect
add channel-wise softmax only (ablation)ConvNeXt-Tiny · CARStop-1 accuracy (%): 87.8 -> 48.9Hurt
use m=1 prototype per class instead of larger m (ablation)OPAL · five fine-grained benchmarks (avg)top-1 accuracy: -1.04 points vs baselineHurt
use m in {3,5,10} prototypes per class (ablation)OPAL · five fine-grained benchmarks (avg)top-1 accuracy: +1.25 points vs baselineImproved
add OPAL to ResNet-50ResNet-50 · CARStop-1 accuracy (%): 71.9 -> 84.3Improved

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