AI Paper Atlas

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

CURE: Consistency-under-Unified Semantic Regularization for Generalized Category Discovery

Yuwei Bian, Shidong Wang, Haofeng Zhang

ICML 2026 · Label-efficient and robust learning › Recognition and long-tail · approach: Contrastive representation theory

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 FT, RSE and LSKL components to the baseline setup (ablation)CURE · CUB-200ACC (All): 70.4 -> 75.2Improved
add FT, RSE and LSKL components (ablation)CURE · CIFAR-100ACC (All): 81.2 -> 85.2Improved
add supervised fine-tuning with prototype alignment in stage I (ablation)CURE · CIFAR-100ACC (New): 72.5 -> 80.0Improved
use hierarchical clustering instead of K-means for centroids (ablation)CURE · CUB-200ACC (All): 73.4 -> 75.2Improved
use spectral instead of hierarchical clustering (ablation)CURE · CUB-200 vs CIFAR-100ACC (All): 75.2 -> 76.7 on CUB; 85.7 -> 85.6 on CIFAR-100Mixed
increase virtual sampling multiplier t to 1.3 (ablation)CURE · CUB-200Old-New gap: 5.6% -> 0.1%Improved
increase virtual sampling multiplier t to 1.2 (ablation)CURE · CIFAR-100Old-New gap: 7.3% -> 1.3%Improved
lower energy threshold epsilon from 0.90 to 0.80 (ablation)CURE · CIFAR-100ACC (All): 85.5 -> 84.1Hurt
set epsilon 0.95 instead of 1.00 (ablation)CURE · CUB-200ACC (All): 73.6 -> 74.6Improved
add stage II semantic expansion on top of stage I (ablation)CURE · ImageNet-100ACC (All): 83.2 -> 87.0Improved
use DINOv2 backbone instead of DINOCURE · CIFAR-100ACC (All): 85.7 -> 89.1Improved
add the symmetric divergence consistency loss (ablation)CURE · CUB-200 and CIFAR-100ACCImproved

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