Browse State-of-the-Art › Semi-Supervised Image Classification

Semi-Supervised Image Classification

130 papers with code · 58 benchmarks · 14 datasets archive 2025-07-28

Computer Vision

Semi-supervised image classification leverages unlabelled data as well as labelled data to increase classification performance.

You may want to read some blog posts to get an overview before reading the papers and checking the leaderboards:

( Image credit: Self-Supervised Semi-Supervised Learning )

Description from the archive archive 2025-07-28.

Benchmarks archive 2025-07-28

60 leaderboard tables shown for this task, 58 with rows (a “benchmark” on this site is a table with at least one row, as on /sota), ordered by row count. “Best model” is the first row in the archive's own order at snapshot; nothing is re-ranked here and metric direction is not recorded in the archive. PwC's Trend sparklines are not in the archive, so that column is omitted. 10 shown of 60 until expanded.

Syntology 2 newer papers were judged to report on this task's benchmarks so far (Claude Sonnet 4.5, reading their tables: a model's judgement, not a result; 0 placed): CIFAR-100, 400 Labels (1 judged, 0 placed), CIFAR-100, 2500 Labels (1 judged, 0 placed), CIFAR-10, 1000 Labels (1 judged, 0 placed). Drawn from 8,886 of 18,637 newer papers checked so far (733 more had nothing to place; 41 could not be read).

DatasetBest model (first row in archive order)PaperCodeSyntologyCompare
ImageNet - 10% labeled data (75 rows) DHO (ViT-Large) Simple Semi-supervised Knowledge Distillation from Vision-Language... code — Compare
ImageNet - 1% labeled data (65 rows) DHO (ViT-Large) Simple Semi-supervised Knowledge Distillation from Vision-Language... code — Compare
CIFAR-10, 4000 Labels (49 rows) Semi-SST (ViT-Small) SST: Self-training with Self-adaptive Thresholding for... — — Compare
cifar-100, 10000 Labels (29 rows) Semi-SST (ViT-Small) SST: Self-training with Self-adaptive Thresholding for... — — Compare
CIFAR-10, 250 Labels (27 rows) Semi-SST (ViT-Small) SST: Self-training with Self-adaptive Thresholding for... — — Compare
CIFAR-10, 40 Labels (21 rows) SemiOccam ViTSGMM: A Robust Semi-Supervised Image Recognition Network Using... code — Compare
CIFAR-100, 400 Labels (21 rows) SemiReward SemiReward: A General Reward Model for Semi-supervised Learning code Syntology official (archive's flag): 4 ran · 4 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 2 where Syntology's instrument failed) · 1 unverified Compare
SVHN, 1000 labels (17 rows) Meta Pseudo Labels (WRN-28-2) Meta Pseudo Labels code Syntology community repositories only · 12 ran (of which 0 constructed an object rather than computing a result; 7 with no instrument failure: 0 honoured, 0 violated, 7 with no contract checked; 5 where Syntology's instrument failed) · 2 unverified Compare
CIFAR-100, 2500 Labels (16 rows) Semi-SST (ViT-Small) SST: Self-training with Self-adaptive Thresholding for... — — Compare
SVHN, 250 Labels (15 rows) ShrinkMatch Shrinking Class Space for Enhanced Certainty in Semi-Supervised Learning code Syntology official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified Compare
STL-10, 1000 Labels (13 rows) Diff-SySC Diff-SySC: An Approach Using Diffusion Models for Semi-Supervised... — — Compare
CIFAR-10, 1000 Labels (9 rows) MixMatch MixMatch: A Holistic Approach to Semi-Supervised Learning code Syntology community repositories only · 47 ran (of which 8 constructed an object rather than computing a result; 27 with no instrument failure: 8 honoured, 0 violated, 19 with no contract checked; 20 where Syntology's instrument failed) · 16 unverified Compare
SVHN, 500 Labels (6 rows) Triple-GAN-V2 (CNN-13) Triple Generative Adversarial Networks code Syntology official (archive's flag): 5 ran · 5 ran (of which 0 constructed an object rather than computing a result; 5 with no instrument failure: 1 honoured, 0 violated, 4 with no contract checked; 0 where Syntology's instrument failed) · 3 unverified Compare
SVHN, 40 Labels (5 rows) ShrinkMatch Shrinking Class Space for Enhanced Certainty in Semi-Supervised Learning code Syntology official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 1 unverified Compare
cifar10, 250 Labels (4 rows) ReMixMatch ReMixMatch: Semi-Supervised Learning with Distribution Alignment... code Syntology official: harvested, nothing ran · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 3 unverified Compare
CIFAR-10, 2000 Labels (4 rows) MixMatch MixMatch: A Holistic Approach to Semi-Supervised Learning code Syntology community repositories only · 47 ran (of which 8 constructed an object rather than computing a result; 27 with no instrument failure: 8 honoured, 0 violated, 19 with no contract checked; 20 where Syntology's instrument failed) · 16 unverified Compare
CIFAR-10, 50 Labels (OpenSet, 6/4) (4 rows) UnMixMatch Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data code — Compare
CIFAR-10, 100 Labels (OpenSet, 6/4) (4 rows) UnMixMatch Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data code — Compare
CIFAR-10, 400 Labels (OpenSet, 6/4) (4 rows) UnMixMatch Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data code — Compare
Mini-ImageNet, 4000 Labels (4 rows) SimPLE SimPLE: Similar Pseudo Label Exploitation for Semi-Supervised... code Syntology official (archive's flag): 3 ran · 3 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 1 honoured, 0 violated, 1 with no contract checked; 1 where Syntology's instrument failed) · 2 unverified Compare
STL-10, 40 Labels (4 rows) SemiOccam ViTSGMM: A Robust Semi-Supervised Image Recognition Network Using... code — Compare
ImageNet - 0.2% labeled data (3 rows) DebiasPL (ResNet-50) Debiased Learning from Naturally Imbalanced Pseudo-Labels code Syntology official (archive's flag): 1 ran · 1 ran (of which 0 constructed an object rather than computing a result; 0 with no instrument failure: 0 honoured, 0 violated, 0 with no contract checked; 1 where Syntology's instrument failed) · 0 unverified Compare
CIFAR-10, 20 Labels (3 rows) MutexMatch (k=0.6C) MutexMatch: Semi-Supervised Learning with Mutex-Based Consistency... code — Compare
cifar-10, 10 Labels (3 rows) BOSS Building One-Shot Semi-supervised (BOSS) Learning up to Fully... code — Compare
Mini-ImageNet, 10000 Labels (3 rows) FeatMatch FeatMatch: Feature-Based Augmentation for Semi-Supervised Learning code Syntology 1 ran (of which 0 constructed an object rather than computing a result; 1 with no instrument failure: 1 honoured, 0 violated, 0 with no contract checked; 0 where Syntology's instrument failed) · 0 unverified Compare
Mini-ImageNet, 1000 Labels (3 rows) MutexMatch MutexMatch: Semi-Supervised Learning with Mutex-Based Consistency... code — Compare
STL-10 (3 rows) EnAET EnAET: A Self-Trained framework for Semi-Supervised and Supervised... code — Compare
CIFAR-10, 80 Labels (2 rows) MutexMatch (k=0.6C) MutexMatch: Semi-Supervised Learning with Mutex-Based Consistency... code — Compare
CIFAR-100 (400 Labels, ImageNet-100 Unlabeled) (2 rows) UnMixMatch Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data code — Compare
CIFAR-100 (250 Labels, ImageNet-100 Unlabeled) (2 rows) CCSSL Class-Aware Contrastive Semi-Supervised Learning code Syntology official (archive's flag): 2 ran · 2 ran (of which 1 constructed an object rather than computing a result; 2 with no instrument failure: 0 honoured, 0 violated, 2 with no contract checked; 0 where Syntology's instrument failed) · 1 unverified Compare
CIFAR-100 (10000 Labels, ImageNet-100 Unlabeled) (2 rows) UnMixMatch Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data code — Compare
CIFAR-10 (250 Labels, ImageNet-100 Unlabeled) (2 rows) UnMixMatch Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data code — Compare
CIFAR-10 (4000 Labels, ImageNet-100 Unlabeled) (2 rows) UnMixMatch Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data code — Compare
CIFAR-100, 5000Labels (2 rows) LiDAM LiDAM: Semi-Supervised Learning with Localized Domain Adaptation... — — Compare
STL-10 (1000 Labels, ImageNet-100 Unlabeled) (2 rows) UnMixMatch Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data code — Compare
SVHN (250 Labels, ImageNet-100 Unlabeled) (2 rows) UnMixMatch Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data code — Compare
SVHN (40 Labels, ImageNet-100 Unlabeled) (2 rows) UnMixMatch Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data code — Compare
SVHN (1000 Labels, ImageNet-100 Unlabeled) (2 rows) UnMixMatch Scaling Up Semi-supervised Learning with Unconstrained Unlabelled Data code — Compare
Caltech-256, 1024 Labels (1 row) UL-Hopfield (ULH) Unsupervised Learning using Pretrained CNN and Associative Memory Bank — — Compare
Caltech-256 (1 row) UL-Hopfield (ULH) Unsupervised Learning using Pretrained CNN and Associative Memory Bank — — Compare
Salinas (1 row) Res-CP Semi-Supervised Hyperspectral Image Classification Using a... code — Compare
EuroSAT, 20 Labels (1 row) SimCLR-kmediods-PAWS Cold PAWS: Unsupervised class discovery and addressing the... code — Compare
Imagenette, 20 Labels (1 row) SimCLR-kmediods-PAWS Cold PAWS: Unsupervised class discovery and addressing the... code — Compare
Imagenette, 100 Labels (1 row) SimCLR-kmediods-PAWS Cold PAWS: Unsupervised class discovery and addressing the... code — Compare
EuroSAT, 100 Labels (1 row) SimCLR-kmediods-PAWS Cold PAWS: Unsupervised class discovery and addressing the... code — Compare
DeepWeeds, 99 Labels (1 row) SimCLR-kmediods-finetuned Cold PAWS: Unsupervised class discovery and addressing the... code — Compare
Caltech-101 (1 row) UL-Hopfield (ULH) Unsupervised Learning using Pretrained CNN and Associative Memory Bank — — Compare
Caltech-101, 202 Labels (1 row) UL-Hopfield (ULH) Unsupervised Learning using Pretrained CNN and Associative Memory Bank — — Compare
CIFAR-10, 500 Labels (1 row) MixMatch MixMatch: A Holistic Approach to Semi-Supervised Learning code Syntology community repositories only · 47 ran (of which 8 constructed an object rather than computing a result; 27 with no instrument failure: 8 honoured, 0 violated, 19 with no contract checked; 20 where Syntology's instrument failed) · 16 unverified Compare
CIFAR-100, 4000 Labels (1 row) UPS (CNN-13) In Defense of Pseudo-Labeling: An Uncertainty-Aware Pseudo-label... code Syntology official: harvested, nothing ran · 0 ran · 5 unverified Compare
CIFAR-100, 5000 Labels (1 row) LiDAM LiDAM: Semi-Supervised Learning with Localized Domain Adaptation... — — Compare
CIFAR-100, 200 Labels (1 row) MutexMatch (k=0.6C) MutexMatch: Semi-Supervised Learning with Mutex-Based Consistency... code — Compare
CIFAR-10, 100 Labels (1 row) SimCLR-kmediods-PAWS Cold PAWS: Unsupervised class discovery and addressing the... code — Compare
CIFAR-10, 30 Labels (1 row) SimCLR-kmediods-PAWS Cold PAWS: Unsupervised class discovery and addressing the... code — Compare
CIFAR-100, 1000 Labels (1 row) EnAET EnAET: A Self-Trained framework for Semi-Supervised and Supervised... code — Compare
STL-10, 5000 Labels (1 row) MixMatch MixMatch: A Holistic Approach to Semi-Supervised Learning code Syntology community repositories only · 47 ran (of which 8 constructed an object rather than computing a result; 27 with no instrument failure: 8 honoured, 0 violated, 19 with no contract checked; 20 where Syntology's instrument failed) · 16 unverified Compare
SVHN, 2000 Labels (1 row) MixMatch MixMatch: A Holistic Approach to Semi-Supervised Learning code Syntology community repositories only · 47 ran (of which 8 constructed an object rather than computing a result; 27 with no instrument failure: 8 honoured, 0 violated, 19 with no contract checked; 20 where Syntology's instrument failed) · 16 unverified Compare
SVHN, 4000 Labels (1 row) MixMatch MixMatch: A Holistic Approach to Semi-Supervised Learning code Syntology community repositories only · 47 ran (of which 8 constructed an object rather than computing a result; 27 with no instrument failure: 8 honoured, 0 violated, 19 with no contract checked; 20 where Syntology's instrument failed) · 16 unverified Compare
ImageNet (0 rows) no rows in the archive — —
CIFAR-10, 40 Labels (0 rows) no rows in the archive — —

Syntology column: samples harvested from the paper's repositories and executed on synthesized fixtures; “ran” is not a correctness claim and does not order the table. Each line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the instrument figure counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the row's paper, the line starts with that repository's state (“official” is the archive's flag, not a verdict on who wrote the code; “community repositories only” when every sample that ran came from a community repository, “official: no sample here; runs from other or unrecorded repositories” when some came from a repository the paper names or has in its text, or from none recorded); hover it for the repositories the samples that ran came from. A dash means no Syntology record for that paper, not a recorded non-run. Each “ran” line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Libraries

Not in the archive: the export carries no per-task library table, so there is nothing to show at snapshot 2025-07-28.

Datasets archive 2025-07-28

14 datasets whose archive record lists this task, ordered by the archive's paper count.

Subtasks archive 2025-07-28

2 subtasks in the archive's task tree.

Parent tasks archive 2025-07-28

Papers archive 2025-07-28

archive papers tagged: 167 · with a code link: 130 · where Syntology ran a sample: 61 (50 with a run with no instrument failure, 11 where every run was a failure of Syntology's instrument) Syntology

Show: all tagged papersonly where code ran (61 of 167 tagged: 50 with a run with no instrument failure, 11 where every run was a failure of Syntology's instrument)

Page 1 of 2: papers 1 to 100 of 167, in archive order: by repositories listed in the archive (most first), then newest first, not by stars (the archive holds no stars, so PwC's “Social” and “Latest” sorts cannot be reproduced). Papers that list no repository come after every paper that lists one.

Papers without a page here are shown as plain text. A Syntology line reads “N ran (of which C constructed an object rather than computing a result; K with no instrument failure: H honoured, V violated, P with no contract checked; I where Syntology's instrument failed) · U unverified”; the figure “where Syntology's instrument failed” counts failures of Syntology's instrument, not of the code. When the archive marks a repository official for the paper, the line starts with that repository's state (the archive's flag, not a verdict on who wrote the code); hover it for the repositories the samples that ran came from. The first sentence of the abstract is shown for the first 30 papers; every abstract is on its paper's page.

Syntology lines on 62 of the papers shown; no Syntology record for the others (a paper without an arXiv id cannot be joined to the graph, and absence from the graph layer is not a recorded non-run). “Ran” means the sample executed on a synthesized fixture, not that the paper's result was reproduced; each line links to that paper's sample list. Syntology's record for this page last changed ; when this build read Syntology's graph is in the build record. For agents: get_harvested_code_for_paper(arxiv_id) lists each paper's samples; how to connect.

Report a problem or propose a change · a person checks every report against the paper or source before anything changes; decisions are listed on /corrections