SOTAVerified

Novel Class Discovery

The goal of Novel Class Discovery (NCD) is to identify new classes in unlabeled data, by exploiting prior knowledge from known classes. In this specific setup, the data is split in two sets. The first is a labeled set containing known classes and the second is an unlabeled set containing unknown classes that must be discovered.

Papers

Showing 51–60 of 65 papers

TitleStatusHype
Towards Novel Class Discovery: A Study in Novel Skin Lesions Clustering—0
TV100: A TV Series Dataset that Pre-Trained CLIP Has Not Seen—0
Uncertainty-guided Open-Set Source-Free Unsupervised Domain Adaptation with Target-private Class Segregation—0
A Closer Look at Novel Class Discovery from the Labeled Set—0
OpenMix: Reviving Known Knowledge for Discovering Novel Visual Categories in An Open World—0
Novel Class Discovery for Long-tailed RecognitionCode0
A Method for Discovering Novel Classes in Tabular DataCode0
An Interactive Interface for Novel Class Discovery in Tabular DataCode0
Découvrir de nouvelles classes dans des données tabulairesCode0
Supervised Knowledge May Hurt Novel Class Discovery PerformanceCode0
Show:102550
← PrevPage 6 of 7Next →

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1AutoNovelClustering Accuracy0.92—Unverified
#ModelMetricClaimedVerifiedStatus
1AutoNovelClustering Accuracy0.75—Unverified
#ModelMetricClaimedVerifiedStatus
1AutoNovelClustering Accuracy0.95—Unverified