AI Paper Atlas

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

Label Hierarchy Transition: Delving Into Class Hierarchies to Enhance Deep Classifiers

Renzhen Wang, De Cai, Kaiwen Xiao, Xixi Jia, Xiao Han, Deyu Meng

TPAMI 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Partial and multi-label learning · 8 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 hierarchical cross-entropy loss with per-level branches (ablation)ResNet-50 · CUB-200-2011 Species (R=90%)ACC: 55.9 -> 58.3Improved
add transition networks (ablation)ResNet-50 + HCL · CUB-200-2011 Species (R=90%)ACC: 58.3 -> 66.7Improved
add confusion loss (ablation)ResNet-50 + HCL + Transition Networks · CUB-200-2011 Species (R=90%)ACC: 66.7 -> 68.8Improved
learn transition matrices via transition networks instead of trainable parameters (ablation)LHT · CUB-200-2011 Species (R=90%)ACC: 58.1 -> 68.8Improved
vary hyper-parameter s of transition network (ablation)LHT · CUB-200-2011ACCNo clear effect
increase confusion loss weight lambda from 0 to 0.01 (ablation)LHT · CUB-200-2011ACCImproved
use a much larger confusion loss weight lambda (ablation)LHT · CUB-200-2011 semi-supervisedACCHurt
apply LHT to ConvNeXt-BaseConvNeXt-Base baseline · CUB-200-2011 Species (R=90%)ACC: 73.3 -> 80.1Improved
apply LHT to Swin-BaseSwin-Base baseline · CUB-200-2011 Species (R=90%)ACC: 72.4 -> 78.9Improved
apply LHT when two finest levels lack labelsHD-CNN baseline · CUB-200-2011 Species (R^2=90%)ACC: 51.0 -> 60.4Improved
apply LHT with fully supervised labelsHRN · CUB-200-2011 Species (R=0%)ACC: 86.7 -> 87.0Improved
apply LHT at heavy label missingbest baseline · Stanford Cars Maker (R=90%)ACC: 59.7 -> 71.4Improved

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