Image Classification On Clothing1M

Metrics

Accuracy

Results

Performance results of various models on this benchmark

Model Name	Accuracy	Paper Title	Repository
DY	71%	Unsupervised Label Noise Modeling and Loss Correction	-
CC	75.4%	Centrality and Consistency: Two-Stage Clean Samples Identification for Learning with Instance-Dependent Noisy Labels	-
Knockoffs-SPR	75.20%	Knockoffs-SPR: Clean Sample Selection in Learning with Noisy Labels	-
MFRW	75.35%	Learning advisor networks for noisy image classification	-
LongReMix	74.38%	LongReMix: Robust Learning with High Confidence Samples in a Noisy Label Environment	-
CCE (SimCLR)	73.27%	Contrastive Learning Improves Model Robustness Under Label Noise	-
MLNT	73.47%	Learning to Learn from Noisy Labeled Data	-
MAE (SimCLR)	73.36%	Contrastive Learning Improves Model Robustness Under Label Noise	-
CoT	70.15%	Co-teaching: Robust Training of Deep Neural Networks with Extremely Noisy Labels	-
SPR	71.16%	Scalable Penalized Regression for Noise Detection in Learning with Noisy Labels	-
FINE + DivideMix	74.37%	FINE Samples for Learning with Noisy Labels	-
Jigsaw-ViT+NCT	75.4%	Jigsaw-ViT: Learning Jigsaw Puzzles in Vision Transformer	-
LRA-diffusion (CC)	75.7%	Label-Retrieval-Augmented Diffusion Models for Learning from Noisy Labels	-
IMAE	73.2%	IMAE for Noise-Robust Learning: Mean Absolute Error Does Not Treat Examples Equally and Gradient Magnitude's Variance Matters	-
MW-Net	73.72%	Meta-Weight-Net: Learning an Explicit Mapping For Sample Weighting	-
ELR+	74.81%	Early-Learning Regularization Prevents Memorization of Noisy Labels	-
Robust f-divergence	73.09%	When Optimizing $f$-divergence is Robust with Label Noise	-
JoCoR	70.3%	Combating noisy labels by agreement: A joint training method with co-regularization	-
LRT	71.74%	Error-Bounded Correction of Noisy Labels	-
CORES2	73.24%	Learning with Instance-Dependent Label Noise: A Sample Sieve Approach	-

0 of 50 row(s) selected.