SOTAVerified

Zero-Shot Image Classification

Zero-shot image classification is a technique in computer vision where a model can classify images into categories that were not present during training. This is achieved by leveraging semantic information about the categories, such as textual descriptions or relationships between classes.

Papers

Recently AddedMost HypedMost ActiveNeeds VerificationMost Verified

Showing 81–90 of 111 papers

TitleStatusHype
Efficient Model-Agnostic Multi-Group Equivariant Networks—0
Noise-Tolerant Few-Shot Unsupervised Adapter for Vision-Language Models—0
GrowCLIP: Data-aware Automatic Model Growing for Large-scale Contrastive Language-Image Pre-training—0
Semantically-Prompted Language Models Improve Visual Descriptions—0
Learning from Children: Improving Image-Caption Pretraining via CurriculumCode0
Text-to-Image Diffusion Models are Zero-Shot ClassifiersCode0
Language-Driven Anchors for Zero-Shot Adversarial RobustnessCode0
Vision-Language Models Performing Zero-Shot Tasks Exhibit Gender-based Disparities—0
RA-CLIP: Retrieval Augmented Contrastive Language-Image Pre-Training—0
DiRaC-I: Identifying Diverse and Rare Training Classes for Zero-Shot Learning—0
Show:102550
← PrevPage 9 of 12Next →
All datasetsCountry211ICinWODinW

Benchmark Results

#ModelMetricClaimedVerifiedStatus
1OpenClip H/14 (34B)(Laion2B)Top-1 accuracy30.01—Unverified
#ModelMetricClaimedVerifiedStatus
1CLIP (ViT B-32)Average Score56.64—Unverified
#ModelMetricClaimedVerifiedStatus
1GLIP (Tiny A)Average Score11.4—Unverified