Home News Latest Papers Tutorials Datasets Wiki SOTA LLM Models GPU Leaderboard Events

English

Learning With Noisy Labels On Animal

Metrics

Accuracy

ImageNet Pretrained

Network

Results

Performance results of various models on this benchmark

Model Name	Accuracy	ImageNet Pretrained	Network	Paper Title	Repository
Jigsaw-ViT	89.0	NO	DeiT-S	Jigsaw-ViT: Learning Jigsaw Puzzles in Vision Transformer
Cross Entropy	79.4	NO	Vgg19-BN	Learning with Feature-Dependent Label Noise: A Progressive Approach
Nested Dropout	81.3	NO	Vgg19-BN	Boosting Co-teaching with Compression Regularization for Label Noise
PLC	83.4	NO	Vgg19-BN	Learning with Feature-Dependent Label Noise: A Progressive Approach
C2MT	85.9	NO	Vgg-19-BN	Cross-to-merge training with class balance strategy for learning with noisy labels
CE + Dropout	81.3	NO	Vgg19-BN	Boosting Co-teaching with Compression Regularization for Label Noise
SURE	89.0	NO	Vgg19-BN	SURE: SUrvey REcipes for building reliable and robust deep networks
SELFIE	81.8	NO	Vgg19-BN	SELFIE: Refurbishing Unclean Samples for Robust Deep Learning
PSSCL	88.74	NO	Vgg19-BN	PSSCL: A progressive sample selection framework with contrastive loss designed for noisy labels
Dynamic Loss	86.5	NO	Vgg19-BN	Dynamic Loss For Robust Learning
InstanceGM	84.6	NO	Vgg19-BN	Instance-Dependent Noisy Label Learning via Graphical Modelling
SSR	88.5	NO	Vgg19-BN	SSR: An Efficient and Robust Framework for Learning with Unknown Label Noise
InstanceGM with ConvNeXt	84.7	NO	ConvNeXt	Instance-Dependent Noisy Label Learning via Graphical Modelling
Nested+Co-teaching (NCT)	84.1	NO	Vgg19-BN	Compressing Features for Learning with Noisy Labels
BtR	88.5	NO	Vgg19-BN	Bootstrapping the Relationship Between Images and Their Clean and Noisy Labels
SPR	86.8	NO	VGG19-BN	Scalable Penalized Regression for Noise Detection in Learning with Noisy Labels
GNL	85.9	NO	Vgg-19-BN	Partial Label Supervision for Agnostic Generative Noisy Label Learning
InstanceGM with ResNet	82.3	NO	ResNet	Instance-Dependent Noisy Label Learning via Graphical Modelling

0 of 18 row(s) selected.