C-Score, a compact framework that evaluates training behavior in three complementary spaces: prediction, feature representation, and optimization, suggests that clean accuracy alone is insufficient for evaluating SSL robustness in open-world environments, and that internal diagnostic signals are necessary for more reliable robustness assessment under unlabeled contamination.
Abstract
Pseudo-label-based semi-supervised learning has achieved strong performance due to its simplicity and scalability. However, it is typically developed under a closed-world assumption that unlabeled data are drawn from the same distribution as labeled data. In practical deployment, unlabeled data are often collected from open environments and may contain OOD samples. Under such contamination, OOD samples may still receive high-confidence predictions and be incorporated into training as if they were valid target examples. This creates an important evaluation problem: clean in-distribution test accuracy may appear stable even when the internal learning dynamics of SSL have already deteriorated. To address this issue, we study hidden collapse in pseudo-label-based SSL under open-world unlabeled contamination from a diagnostic evaluation perspective. We present C-Score, a compact framework that evaluates training behavior in three complementary spaces: prediction, feature representation, and optimization. C-Score includes PLE and CCI for unlabeled prediction behavior, Sem-Drift for deviation from labeled semantic anchors, and Grad-Align for the compatibility between labeled and unlabeled optimization. Experiments on CIFAR-10 and CIFAR-100 with multiple OOD sources, varying contamination ratios, and four pseudo-label-based SSL algorithms show that C-Score metrics reveal hidden degradation that clean accuracy alone fails to detect: under SVHN contamination, CCI rises over 280% while best-accuracy remains within 3% of the uncontaminated baseline; near-OOD sources (CIFAR-100, STL-10) cause up to 14.9% accuracy collapse (FlexMatch, r=0.5). The results suggest that clean accuracy alone is insufficient for evaluating SSL robustness in open-world environments, and that internal diagnostic signals are necessary for more reliable robustness assessment under unlabeled contamination.
DPCL was leveraged to learn better representation distributions by uniformly projecting data on a hypersphere, and by which more accurate pseudo labels could be generated, and achieves consistent improvements over existing state-of-the-art methods across multiple benchmarks.
Pengfei Lv, Jing Chai· International Conference on...· 0 citations
Self-supervised learning (SSL) has emerged as a promising approach for tabular data, yet its efficacy under extreme label scarcity and test-time missingness remains under-explored. In this paper, we evaluate a mask-and-recover SSL pretraining objective against training from scratch and classical baselines across 14 diverse classification tasks. First, while SSL outperforms training from scratch on average and remains competitive with state-of-the-art tree ensembles (achieving ~0.8954 AUC vs. Random Forest's 0.9015 at 10% labels), the SSL-vs-scratch gains exhibit high inter-task variance and lack significance (p = 0.626 at both 5% and 10% labels). Second, contrary to the hypothesis that missing-value imputation objectives universally benefit datasets with native missingness, SSL yields the most reliable improvements on clean datasets, while frequently degrading performance on datasets with high inherent missingness. Third, despite this training variance, SSL-pretrained models achieve a higher average AUC than scratch-trained models under both test-time missingness completely at random (MCAR) injection (+0.0245 AUC, positive on 11 of 14 tasks) and structured missingness shifts (MNAR, +0.0418 AUC, positive on 8 of 14 tasks), though neither difference remains statistically significant after Holm-Bonferroni correction for multiple comparisons (adjusted p = 0.118 and p = 0.518, respectively). Fourth, comparing our mask-and-recover objective against three established tabular SSL baselines (VIME, SCARF, SubTab) under an identical encoder architecture, we find no significant difference from any of them (adjusted p = 0.459, p = 1.000, p = 1.000), indicating our findings reflect general properties of tabular SSL rather than idiosyncrasies of one particular pretext task.
Self-supervised learning matches supervised accuracy from a fraction of the labels, but the labeled-sample efficiency behind this has lacked a theoretical explanation. We provide one. Data augmentation induces a similarity graph on the unlabeled data, so downstream learning on that graph is graph-Laplacian-regularized learning. We prove a fast transductive rate, $O(1/n_L)$ in the number of labels, in place of the supervised $O(1/\sqrt{n_L})$, by carrying the leave-one-out stability apparatus of Johnson and Zhang (JMLR 2007) over to the augmentation graph, and without the unrealistic assumptions of limit-based analyses (exact kernel, generalizing features). The bound makes augmentation quality explicit: the expected error is at most $C/n_L + R_{\mathrm{DA}}(y)$, where the data-augmentation alignment error $R_{\mathrm{DA}}(y)$ is proportional to the graph-cut mass of augmentations that cross a label boundary, so good augmentations let few labels suffice. The analysis uses a streamlined loss that drops the projector, negative-sample, and orthogonality overhead of standard objectives yet still recovers the top-$K$ ideal features in the infinite-data limit, the augmentation-kernel eigenspace studied by Zhai et al. The bound gives a mechanistic account of the accuracy-versus-label-count curve through augmentation quality, verified in a controlled model where the constants are known.
Existing semi-supervised learning (SSL) methods rely predominantly on pseudo-labeling and consistency regularization to leverage unlabeled data, demonstrating significant performance improvements. However, we pinpoint that these methods suffer from a confidence-for-weighting issue, overvaluing high-confidence pseudo-labels while undervaluing low-confidence yet informative samples that are critical for robust generalization. In this paper, we introduce EntropyMatch, an entropy-driven SSL framework that redefines sample importance through prediction entropy rather than confidence alone. EntropyMatch employs a bidirectional weighting strategy: upward exploitation exploits reliable hard samples to refine decision boundaries while downward exploration cautiously explores uncertain ones to reduce noise. Additionally, EntropyMatch features an adaptive training mechanism that aligns with model maturity, shifting focus from safe exploration to strategic exploitation as training progresses. Experiments on eight benchmarks across various SSL tasks—spanning image classification, facial expression recognition, and human action recognition—validate EntropyMatch’s robustness and effectiveness. It consistently achieves state-of-the-art results, notably matching state-of-the-art LION’s performance on RAF-DB with just half the labeled data, demonstrating superior data efficiency and generalization.
Anyang Tong, Zenglin Shi, Zhun Zhong et al.· IEEE Transactions on Image P...· 0 citations
Early stopping is a fundamental technique for mitigating overfitting in machine learning and is especially critical in many weakly supervised learning scenarios. Conventional early stopping, however, needs a sufficiently sized and representative validation set. This creates a trade-off: dedicating more data for validation can diminish the training set and limit model performance, while an insufficient validation set may lead to sub-optimal model selection. This challenge is particularly serious in many weakly supervised learning scenarios where validation data may be entirely unavailable or unreliable. In this paper, we propose Label Wave, which does not require validation data for selecting the desired model across various weakly supervised learning paradigms, including learning with noisy labels (LNL), positive-unlabeled learning, and unlabeled-unlabeled learning. It works by tracking the changes in the model's predictions on the training set during the training process, aiming to halt training before the model excessively fits misleading patterns. This method is empirically supported by our observation that minimum fluctuations in predictions typically occur at the training epoch before the model transitions from fitting reliable patterns to misleading ones. Extensive experiments further provide large-scale evaluation for LNL across diverse noise conditions, model families, optimizers, and data modalities. Label Wave selects near-oracle checkpoints (test-accuracy oracle gaps: 0.15-0.93 percentage points in controlled sweeps and 1.90 percentage points on 71 tabular tasks) and boosts 7 LNL methods by 1.98/2.86 percentage points over 10%/20% hold-out validation.
Suqin Yuan, Muyang Li, Lei Feng et al.· IEEE Transactions on Pattern...· 0 citations
Multi-Label Learning (MLL) refers to inducing multi-label prediction models from the precisely labeled training dataset. However, in many real-world scenarios, e.g., crowdsourcing annotations, the training datasets are often only partially valid, where each training instance is associated with a candidate label set, covering ground-truth labels but also with irrelevant ones. Naturally, learning with such datasets, formally referred to as Partial Multi-label Learning (PML), involves many noisy supervised signals, hence imposing a significant challenge to the prediction model induction. To meet this challenge, we purify the noisy supervised signals by formulating the latent label distribution, i.e., the probability of a candidate label being a ground-truth one, and then jointly learn it with the prediction model by minimizing their regularized Wasserstein distance, i.e., a robust distance for distributions as well as involving label correlations. Therefore, we propose a novel PML method, namely Wasserstein Partial Multi-Label Learning with dual Label Correlation Perspectives (Wpml3cp), solved by the gradient descent with an augmented Lagrange multiplier technique. To further enhance the robustness of Wpml3cp against exceptionally high ratios of irrelevant labels, we extend it with a Dual-branch Competitive Cleansing mechanism, leading to Wpml3cp-D. Besides, we also analyze the generalization error bound and time complexity of Wpml3cp and Wpml3cp-D. The extensive experiments are constructed by comparing Wpml3cp and Wpml3cp-D with existing PML baselines across synthetic and real-world datasets, and empirical results demonstrate that Wpml3cp and Wpml3cp-D can outperform the PML baselines in various noisy levels.
Ximing Li, Yuanchao Dai, Bing Wang et al.· ACM Transactions on Knowledg...· 0 citations