Papers › ViTSGMM: A Robust Semi-Supervised Image Recognition Network Using Sparse Labels

ViTSGMM: A Robust Semi-Supervised Image Recognition Network Using Sparse Labels

4 Jun 2025SSRN Electronic Journal 2025 3arXiv:2506.03582archive 2025-07-28

Rui Yann, Xianglei Xing

We present ViTSGMM, an image recognition network that leverages semi-supervised learning in a highly efficient manner. Existing works often rely on complex training techniques and architectures, while their generalization ability when dealing with extremely limited labeled data remains to be improved. To address these limitations, we construct a hierarchical mixture density classification decision mechanism by optimizing mutual information between feature representations and target classes, compressing redundant information while retaining crucial discriminative components. Experimental results demonstrate that our method achieves state-of-the-art performance on STL-10 and CIFAR-10/100 datasets when using negligible labeled samples. Notably, this paper also reveals a long-overlooked data leakage issue in the STL-10 dataset for semi-supervised learning tasks and removes duplicates to ensure the reliability of experimental results. Code available at https://github.com/Shu1L0n9/ViTSGMM.

PaperPDFConference PDFCode

Code

Shu1L0n9/SemiOccam officialmentioned on GitHubpytorch report

Repository list and official/mentioned flags are the archive's, frozen 2025-07-28. Reachability, where shown, is from one Syntology probe window (2026-09-16 to 2026-09-18); repositories not probed show nothing. GitHub stars are not tracked.

Code Syntology ran Syntology

Not run by Syntology. Nothing on this page verifies that the listed code works.

Tasks

Semi-Supervised Image Classification

Datasets

Introduced by this paper, per the archive.

CleanSTL-10

Results from the paper archive 2025-07-28

TaskDatasetModelMetricValueRank at snapshotLeaderboardReport
Semi-Supervised Image Classification CIFAR-10, 250 Labels SemiOccam Percentage error 3.47 #3 of 27 Archive leaderboard report
Semi-Supervised Image Classification CIFAR-10, 40 Labels SemiOccam Percentage error 3.51 #1 of 21 Archive leaderboard report
Semi-Supervised Image Classification CIFAR-100, 2500 Labels SemiOccam Percentage error 22.19 #3 of 16 Archive leaderboard report
Semi-Supervised Image Classification CIFAR-100, 400 Labels SemiOccam Percentage error 26.59 #3 of 21 Archive leaderboard report
Semi-Supervised Image Classification STL-10, 40 Labels SemiOccam Accuracy 95.43 #1 of 4 Archive leaderboard report

Ranks are positions in the archive's leaderboards as they stood at the 2025-07-28 snapshot. Results published since then are not among these rows, so a rank here is not a current standing.

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