JEPAMatch: Geometric Representation Shaping for Semi-Supervised Learning

TL;DR

This paper introduces JEPAMatch, a semi-supervised learning method that shapes geometric representations in latent space to improve class balance, reduce noise, and accelerate training, outperforming existing methods.

Contribution

It proposes a novel training objective combining semi-supervised loss with latent-space regularization inspired by LeJEPA, enhancing representation quality and training efficiency.

Findings

Outperforms existing semi-supervised methods on CIFAR-100, STL-10, and Tiny-ImageNet.

Accelerates convergence and reduces computational cost compared to FixMatch-based methods.

Produces more balanced and well-structured representations in latent space.

Abstract

Semi-supervised learning has emerged as a powerful paradigm for leveraging large amounts of unlabeled data to improve the performance of machine learning models when labeled data are scarce. Among existing approaches, methods derived from FixMatch have achieved state-of-the-art results in image classification by combining weak and strong data augmentations with confidence-based pseudo-labeling. Despite their strong empirical performance, these methods typically struggle with two critical bottlenecks: majority classes tend to dominate the learning process, which is amplified by incorrect pseudo-labels, leading to biased models. Furthermore, noisy early pseudo-labels prevent the model from forming clear decision boundaries, requiring prolonged training to learn informative representation. In this paper, we introduce a paradigm shift from conventional logical output threshold base, toward an explicit shaping of geometric representations. Our approach is inspired by the recently proposed Latent-Euclidean Joint-Embedding Predictive Architectures (LeJEPA), a theoretically grounded framework asserting that meaningful representations should exhibit an isotropic Gaussian structure in latent space. Building on this principle, we propose a new training objective that combines the classical semi-supervised loss used in FlexMatch, an adaptive extension of FixMatch, with a latent-space regularization term derived from LeJEPA. Our proposed approach, encourages well-structured representations while preserving the advantages of pseudo-labeling strategies. Through extensive experiments on CIFAR-100, STL-10 and Tiny-ImageNet, we demonstrate that the proposed method consistently outperforms existing baselines. In addition, our method significantly accelerates the convergence, drastically reducing the overall computational cost compared to standard FixMatch-based pipelines.