This work introduces a novel framework, Gaussian Bridge Consistency (GBC), to address challenges of semi-supervised learning by constructing semantic interpolation paths between unlabeled samples and high-quality class anchors, and proposes BridgeMix, a confidence-aware feature mixing strategy that interpolates both sample and anchor pairs to amplify cross-sample generalization.
Abstract
Real-world semi-supervised learning (SSL) often encounters significant challenges with long-tailed label distributions and noisy pseudo-labels, which hinder generalization and amplify confirmation bias. In this work, we introduce a novel framework, Gaussian Bridge Consistency (GBC), to address these challenges by constructing semantic interpolation paths between unlabeled samples and high-quality class anchors. Our method maintains a dynamic Prototype Atlas that stores a diverse and evolving set of labeled and pseudo-labeled exemplars per class. For each unlabeled instance, GBC forms a class-conditional Gaussian Feature Bridge in the latent space, enabling the student model to traverse a smooth trajectory from uncertain predictions to reliable class prototypes. A bridge consistency loss is applied along this path to enforce alignment with a geometrically interpolated target distribution. Furthermore, we propose BridgeMix, a confidence-aware feature mixing strategy that interpolates both sample and anchor pairs to amplify cross-sample generalization. Extensive experiments on CIFAR10-LT and ImageNet-LT (USB benchmarks) validate the robustness and effectiveness of GBC under realistic long-tailed SSL settings, consistently improving long tail-class performance without sacrificing scalability.
Modern semi-supervised learning (SSL) couples pseudo-label generation and classifier training, using the classifier's own confidence to select the pseudo-labels that are then used to update the model. In the cold-start regime, where at most a few labels per class are available, this coupling is ill-posed, since the classifier cannot supervise itself before it has learned. To address this problem, we propose VAST (Veracity-Aware Semi-Supervised Training), which decouples these two stages. Probabilistic beliefs over the unlabeled set are first inferred directly from the geometry of a frozen self-supervised embedding and only then distilled into an inductive classifier. The construction rests on the Veracity Matrix, a kernel-based structure that aggregates label evidence across the data manifold and admits an interpretation as a Dirichlet posterior under a per-observation powered-likelihood model. Additionally, we introduce Veracity Propagation, a self-terminating belief-spreading step that extends coverage beyond the kernel neighborhood of the labeled set. Under a controlled protocol in which all methods receive identical frozen embeddings and labeled sets, VAST outperforms the strongest graph-based SSL baselines at every operating point across three datasets, with statistically significant gains in 7 of 9 comparisons, while producing a deployable inductive classifier rather than requiring transductive graph inference. Compared with end-to-end confidence-gated SSL, we further find that these methods underperform in this setting and, in our experiments, do not consistently exceed labeled-only performance.
Semi-supervised Class Incremental Learning (SSCIL) is a severe challenge for neural networks, and it is hardest in the exemplar-free setting where no past data may be stored. Existing methods forget catastrophically due to feature drift, and their pseudo-labels become increasingly unreliable as the label space grows. In this paper, we propose MAGIC (Manifold Anchoring and Geometric Incremental Calibration), a framework that stabilizes plasticity without storing exemplars. MAGIC's design centers on two components. The first is Soft-Weighted Geometry Calibration (SWGC), which uses graph-based label propagation on the learner's plastic feature space to weight and calibrate class means and variances computed on the frozen backbone; from these calibrated Gaussians, we sample phantom features that stand in for data from previous tasks. The second is a Geometric Structural Alignment (GSA) objective that preserves representation topology by matching the relational structure of student and teacher heads and aligning feature anchors with the fixed classifier prototypes, locking the orientation of the feature space. Together, these constraints keep the adapter from drifting, so geometric relations between classes remain stable as new classes arrive. We implement MAGIC with a frozen ResNet-18 backbone and a learnable plastic adapter. Across CIFAR-100, CUB-200, and ImageNet-R, at label ratios of 1%, 5%, and 10%, MAGIC improves average incremental accuracy over most of the supervised CIL methods equipped with FixMatch and native SSCIL baselines; the largest gains occur in the fine-grained, low-label setting, where confidence thresholding fails most clearly.
Yousef Abdi, Mohammad Asadpour, Y. Seyfari· 0 citations
This work proposes a teacher-student semi-supervised learning framework that generates high-quality pseudo-labels from unlabeled data through confidence-aware map refinement, and introduces a spatial clipping technique that selectively preserves high-confidence regions while removing unreliable segments.
Chikao Tsuchiya, Dhaval Bhanderi, David Ilstrup et al.· 0 citations
Semi-supervised learning (SSL) is an effective approach to leverage limited labeled data alongside abundant unlabeled data. While methods like MarginMatch have shown promise by using the average pseudo margin (APM) to evaluate pseudo-label reliability, they face two limitations: (1) APM struggles to distinguish correctly and incorrectly pseudo-labeled classes for hard-to-learn samples, and (2) fixed-percentile thresholding leads to suboptimal pseudo-label utilization throughout training. To address these challenges, we propose APIMatch, a novel SSL framework introducing the average pseudo integrated margin (APIM) metric and a negative-sample-aware dynamic percentile thresholding strategy. The core novelty lies in jointly modeling the pseudo-labeled class and competitive non-pseudo-labeled classes to accurately re-evaluate hard-to-learn samples, coupled with a confidence-distribution-aware adaptive threshold. Specifically, APIM jointly considers the logit differences between the pseudo-labeled class and both the largest and second-largest non-pseudo-labeled classes, enabling more accurate confidence evaluation for hard-to-learn samples. In addition, our dynamic thresholding strategy constructs a pseudo-negative-sample reference set to model confidence distributions and adaptively adjusts the percentile threshold based on training progression, improving both recall in early stages and precision in later stages. Extensive experiments on CIFAR-10, CIFAR-100, and STL-10 demonstrate that APIMatch achieves competitive performance against state-of-the-art methods, particularly in extremely low-label scenarios. Under the 1 label per class setting, APIMatch achieves error rates of
1.84
%
and
36.7
%
on CIFAR-10 and CIFAR-100, respectively. On STL-10 with only 4 labels per class, APIMatch achieves an error rate of
18.47
%
±
0.65
%
, representing a
6.9
%
absolute accuracy improvement over MarginMatch, demonstrating strong robustness under severe label scarcity.
Xianmin Wang, Dong-Yuan Tian, Jing Li· The European Journal on Arti...· 0 citations
Despite recent advances, self-supervised learning (SSL) models and Joint-Embedding Predictive Architectures (JEPAs) remain susceptible to learning spurious biases in the dataset. These techniques rely on regularization, which prevents representation collapse by enforcing a global target distribution such as a multivariate Gaussian or a uniform distribution on the sphere. However, these global constraints are insufficient to prevent bias entanglement, as task-irrelevant features can still segregate the latent space into distinct sub-regions. While recent approaches like Entangling and Disentangling (EnD) and Fair Supervised Contrastive Learning (FSCL) empirically debias the latent space, we show that they act as partial approximations of conditional distribution matching. To enforce this matching explicitly, we propose Unbiased Open World Regularization (UOWReg), an encoder-only framework. We show that this shift from a global to a conditional objective guarantees statistical independence between the learned representations and the targeted attributes, regardless of the chosen target distribution. We empirically validate this framework across both Gaussian and spherical latent spaces, using statistical measures to enforce these target distributions. While conditional matching successfully mitigates bias with both distributions, we demonstrate that enforcing conditional uniformity on the sphere yields a lower linearprobing classification error. Empirically, UOWReg reduces Equalized Odds violations on the CelebA benchmark while maintaining competitive classification accuracy compared to existing encoder-only baselines. Furthermore, we introduce the Synthetic Engraving Task-a novel setting in which a dominant macro-structure masks a fine-grained micro-signature. We show that UOWReg effectively prevents the subpopulation collapse observed in standard SSL, successfully isolating micro-signatures even when heavily entangled with the global structure.
Léo Nicollier, M. Pic, Pablo Musé et al.· arXiv.org· 0 citations
Analytic Continual Learning (ACL) offers a computationally efficient alternative to gradient-based approaches. Recent ACL methods are based on Recursive Least Squares (RLS) and have achieved the state-of-the-art results compared to other alternatives. However, they falter significantly in Class-Incremental Learning scenarios characterized by Long-Tailed distributions. While the ill-conditioning of the autocorrelation (Gram) matrix is a known limitation of RLS, we demonstrate that class imbalance exacerbates this issue into a distinct spectral pathology:"tail"classes suffer from severe spectral collapse, rendering their subspaces numerically indistinguishable from noise. Standard Ridge Regression ($L_2$) fails to address this effectively as it applies isotropic regularization - a uniform penalty that is insufficient to stabilize the tail without over-shrinking the head. To address this, we propose Geometry-Spectral Rectification (GSR), a theoretically grounded framework that treats long-tailed learning as a spectral regularization problem. Unlike standard isotropic regularization (Ridge) which uniformly penalizes all eigenvalues, GSR acts as an anisotropic spectral filter, selectively inflating the collapsed eigenvalues of tail classes. We construct a structured, data-dependent spectral perturbation matrix $\Delta$ that selectively inflates collapsed tail eigen-directions of the Gram matrix. Theoretical analysis proves that GSR guarantees an improved stable rank for the Gram matrix, ensuring numerical stability. Extensive experiments show that GSR establishes a new state-of-the-art for analytic CIL, offering a superior trade-off between computational efficiency and robust generalization in long-tailed settings.
Q. Tran, Ngoc-Hai Nguyen, Quan Dao et al.· arXiv.org· 0 citations
We use cookies to run the site and, with your consent, for analytics and to show ads.
See our Cookie Policy.
