Image Classification on CIFAR-10, 250 Labels

Metric: Percentage error (lower is better)

LeaderboardDataset

Loading chart...

Results

Sort:

#	Model↕	Percentage error▲	Extra Data	Paper	Date↕	Code
1	SemiOccam	3.47	No	ViTSGMM: A Robust Semi-Supervised Image Recognit...	2025-06-04	Code
2	DebiasPL (w/ FixMatch)	4.6	No	Debiased Learning from Naturally Imbalanced Pseu...	2022-01-05	Code
3	ShrinkMatch	4.74	No	Shrinking Class Space for Enhanced Certainty in ...	2023-08-13	Code
4	SimMatch	4.84	No	SimMatch: Semi-supervised Learning with Similari...	2022-03-14	Code
5	NP-Match	4.87	No	NP-Match: When Neural Processes meet Semi-Superv...	2022-07-03	Code
6	FreeMatch	4.88	No	FreeMatch: Self-adaptive Thresholding for Semi-s...	2022-05-15	Code
7	FixMatch+CR	5.04	No	Contrastive Regularization for Semi-Supervised L...	2022-01-17	-
8	ReMixMatch	6.27	No	ReMixMatch: Semi-Supervised Learning with Distri...	2019-11-21	Code
9	EnAET	7.6	No	RealMix: Towards Realistic Semi-Supervised Deep ...	2019-12-18	Code
10	RealMix	9.79	No	RealMix: Towards Realistic Semi-Supervised Deep ...	2019-12-18	Code
11	MixMatch	11.08	No	MixMatch: A Holistic Approach to Semi-Supervised...	2019-05-06	Code
12	LiDAM	19.17	No	LiDAM: Semi-Supervised Learning with Localized D...	2020-10-13	-
13	VAT	36.03	No	Virtual Adversarial Training: A Regularization M...	2017-04-13	Code
14	MeanTeacher	47.32	No	Mean teachers are better role models: Weight-ave...	2017-03-06	Code
15	MixUp	47.43	No	mixup: Beyond Empirical Risk Minimization	2017-10-25	Code
16	Ⅱ-Model	53.12	No	Temporal Ensembling for Semi-Supervised Learning	2016-10-07	Code

#1SemiOccamSOTA
3.47
Percentage error· 2025-06-04
ViTSGMM: A Robust Semi-Supervised Image Recognition Network Using Sparse Labels Code
#2DebiasPL (w/ FixMatch)SOTA
4.6
Percentage error· 2022-01-05
Debiased Learning from Naturally Imbalanced Pseudo-Labels Code
#3ShrinkMatch
4.74
Percentage error· 2023-08-13
Shrinking Class Space for Enhanced Certainty in Semi-Supervised Learning Code
#4SimMatch
4.84
Percentage error· 2022-03-14
SimMatch: Semi-supervised Learning with Similarity Matching Code
#5NP-Match
4.87
Percentage error· 2022-07-03
NP-Match: When Neural Processes meet Semi-Supervised Learning Code
#6FreeMatch
4.88
Percentage error· 2022-05-15
FreeMatch: Self-adaptive Thresholding for Semi-supervised Learning Code
#7FixMatch+CR
5.04
Percentage error· 2022-01-17
Contrastive Regularization for Semi-Supervised Learning
#8ReMixMatchSOTA
6.27
Percentage error· 2019-11-21
ReMixMatch: Semi-Supervised Learning with Distribution Alignment and Augmentation Anchoring Code
#9EnAET
7.6
Percentage error· 2019-12-18
RealMix: Towards Realistic Semi-Supervised Deep Learning Algorithms Code
#10RealMix
9.79
Percentage error· 2019-12-18
RealMix: Towards Realistic Semi-Supervised Deep Learning Algorithms Code
#11MixMatchSOTA
11.08
Percentage error· 2019-05-06
MixMatch: A Holistic Approach to Semi-Supervised Learning Code
#12LiDAM
19.17
Percentage error· 2020-10-13
LiDAM: Semi-Supervised Learning with Localized Domain Adaptation and Iterative Matching
#13VATSOTA
36.03
Percentage error· 2017-04-13
Virtual Adversarial Training: A Regularization Method for Supervised and Semi-Supervised Learning Code
#14MeanTeacherSOTA
47.32
Percentage error· 2017-03-06
Mean teachers are better role models: Weight-averaged consistency targets improve semi-supervised deep learning results Code
#15MixUp
47.43
Percentage error· 2017-10-25
mixup: Beyond Empirical Risk Minimization Code
#16Ⅱ-ModelSOTA
53.12
Percentage error· 2016-10-07
Temporal Ensembling for Semi-Supervised Learning Code