CAT: Class Aware Adaptive Thresholding for Semi-Supervised Domain Generalization

TL;DR

This work tackles semi-supervised domain generalization (SSDG) by introducing CAT, a classifier-aware adaptive thresholding framework that jointly optimizes pseudo-label quality and domain-invariant representation learning. CAT employs global and class-specific adaptive thresholds, updated through EMA, to select pseudo-labels with higher class diversity while incorporating a noisy-label refinement stage via supervised contrastive learning on refined pseudo-labels. The final objective combines supervised loss, unsupervised pseudo-label loss, and supervised contrastive loss, with a fixed weight for the unlabeled data term, achieving strong performance across PACS, OfficeHome, VLCS, and miniDomainNet in low-label regimes. Empirical results show CAT consistently outperforms state-of-the-art SSDG methods, demonstrating improved robustness to domain shifts and practical viability in data-scarce scenarios.

Abstract

Domain Generalization (DG) seeks to transfer knowledge from multiple source domains to unseen target domains, even in the presence of domain shifts. Achieving effective generalization typically requires a large and diverse set of labeled source data to learn robust representations that can generalize to new, unseen domains. However, obtaining such high-quality labeled data is often costly and labor-intensive, limiting the practical applicability of DG. To address this, we investigate a more practical and challenging problem: semi-supervised domain generalization (SSDG) under a label-efficient paradigm. In this paper, we propose a novel method, CAT, which leverages semi-supervised learning with limited labeled data to achieve competitive generalization performance under domain shifts. Our method addresses key limitations of previous approaches, such as reliance on fixed thresholds and sensitivity to noisy pseudo-labels. CAT combines adaptive thresholding with noisy label refinement techniques, creating a straightforward yet highly effective solution for SSDG tasks. Specifically, our approach uses flexible thresholding to generate high-quality pseudo-labels with higher class diversity while refining noisy pseudo-labels to improve their reliability. Extensive experiments across multiple benchmark datasets demonstrate the superior performance of our method, highlighting its effectiveness in achieving robust generalization under domain shift.

Figures (1)

• Figure 1: Comparison between our method with StyleMatch and MultiMatch in different label settings.