AI Paper Atlas

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

Hier-COS: Making Deep Features Hierarchy-aware via Composition of Orthogonal Subspaces

Depanshu Sani, Saket Anand

CVPR 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Learning from unlabeled and positive-unlabeled data · 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 Hier-COS instead of HAFrameResNet-50 · FGVC-Aircrafttop-1 accuracy: 80.55 -> 81.75Improved
use Hier-COS instead of HAFrameResNet-50 · CIFAR-100top-1 accuracy: 77.53 -> 77.79No clear effect
use Hier-COS instead of HAFrameResNet-50 · iNaturalist-19HOPS: 0.89 -> 0.96Improved
use Hier-COS instead of HAFrameResNet-50 · FGVC-AircraftHOPS: 0.86 -> 0.89Improved
use Hier-COS instead of cross-entropy with ViT backboneViT · iNaturalist-19top-1 accuracy: 78.39 -> 80.81Improved
use Hier-COS instead of cross-entropy with ViT backboneViT · iNaturalist-19HOPS: 0.53 -> 0.98Improved
use Hier-COS instead of HAFrameResNet-50 · FGVC-Aircraftmistake severity: 2.00 -> 2.09Hurt
use Hier-COS instead of Flamingo-IResNet-50 · FGVC-AircraftFPA: 77.92 -> 81.56Improved
use Hier-COS instead of HAFeat-IResNet-50 · CIFAR-100FPA: 75.52 -> 77.37Improved
use Hier-COS instead of Flamingo-IIResNet-50 · iNaturalist-19FPA: 68.75 -> 69.93Improved
use Hier-COS instead of Cross Entropy (super-class probabilities summed)ResNet-50 · CIFAR-100FPA: 77.11 -> 77.37Improved

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