Learning on Imbalanced Noisy Data via Debiased Sample Selection and LLM-Driven Annotation

CANNE: CLIP-Based ANNE Selection for Noisy-Label Learning

Learning with noisy labels (LNL) remains challenging, especially when the identification of clean samples relies heavily on the predictions of the model being trained. In such cases, early-stage selection errors may be reinforced during iterative optimization, leading to unreliable supervision. To alleviate this issue, a two-stage framework, termed CANNE, is proposed by combining Contrastive Language–Image Pre-training (CLIP)-based conservative offline cleaning with Adaptive Nearest Neighbors and eigenvector-based sample selection (ANNE)-based online refinement. Specifically, a high-confidence clean seed set is first constructed using two complementary probability sources derived from frozen CLIP representations and reliability criteria, including class-wise loss modeling and prediction consistency. This seed set is then used as a set of reliable anchors during the subsequent ANNE training process, where online feature- and neighborhood-based refinement further recovers and adjusts sample partitions. In this way, CANNE uses external vision–language priors to provide conservative and persistent guidance while preserving the adaptive recovery ability of online noisy-label learning. Experimental results on CIFAR-10, CIFAR-100, Animal-10N, and Mini-WebVision, together with additional evaluation under open-set noise, show that the proposed method achieves competitive performance across diverse noisy-label settings. In particular, CANNE achieves 96.6% and 96.3% best accuracies on CIFAR-10 under 80% and 90% symmetric noise, respectively, and 81.0% and 79.0% on CIFAR-100 under 20% and 50% symmetric noise. Additional repeated-run, threshold-sensitivity, and runtime analyses further indicate that the CLIP-based seed set provides stable guidance with only moderate computational overhead.

SoREL: Soft-Label Refurbishment with Ensemble Learning for Noisy Long-Tailed Classification

Robust Ensemble Learning Under Label Noise: A Theoretical Analysis and Framework-Specific Solutions.

Ensemble learning methods combine multiple weak base learners to create a robust decision model, effectively analyzing feature-response relationships across various domains. However, the assumption of accurate sample-label relationships in supervised learning often breaks down in real-world datasets, leading to performance degradation due to incorrect information. The framework-specific effects of label noise on ensemble learning have not been sufficiently explored from a theoretical perspective. This article investigates the problem of learning from datasets contaminated by label noise within ensemble frameworks. We utilize bias-variance-diversity (BVD) decomposition theory to examine the impact of noisy labels on three mainstream ensemble paradigms: Bagging, Boosting, and Stacking. Our theoretical analysis characterizes the mechanisms behind performance degradation and guides the development of targeted strategies: data filtering (DF) for Bagging, sample reweighting (SR) for Boosting, and interactive feature purification (IFPS) for stacking. We validate our approaches on synthetic and real-world noisy-label benchmarks, demonstrating consistent improvements over traditional ensemble methods through extensive comparison and ablation experiments. Our findings offer actionable insights for enhancing the robustness of ensemble learning in the presence of noisy labels, thereby broadening its applicability in practical scenarios.

Certainty-Aware Partition and Sufficient Utilization of Noisy Samples

Distribution-aware probability contrastive learning for class-imbalanced semi-supervised learning

Semi-Supervised Learning (SSL) has shown significant advantages by leveraging abundant unlabeled data to enhance model performance with successful applications in computer vision. However, existing SSL methods might exhibit significant performance degradation in real applications, primarily due to the learning bias stemming from the following two challenges: (1) the occurrence of class imbalance in real-world datasets, and (2) the misalignment between class distributions of labeled and unlabeled data. To deal with the above challenges, current Class-Imbalanced SSL (CISSL) methods mainly resorted to data rebalancing strategies (e.g., resampling) and failed to fully exploit the high-level distributional characteristics. To address this problem, a novel solution termed Distribution-Aware Probability Contrastive Learning (DPCL) is proposed in this work. In DPCL, contrastive learning was leveraged to learn better representation distributions by uniformly projecting data on a hypersphere, and by which more accurate pseudo labels could be generated. To evaluate the classification performance of DPCL, we conduct extensive experiments on benchmark class-imbalanced SSL datasets. The results demonstrate that DPCL achieves consistent improvements over existing state-of-the-art methods across multiple benchmarks.

PaSta: Noisy Node Classification with Partial Label Learning

Noisy node classification problem is a fundamental yet challenging task for real-world graph-related web services, where node labels are often corrupted or unreliable due to weak supervision or automatic annotation. However, existing methods typically train models based on one-hot labels, which not only makes models susceptible to overfitting on noisy labels, but also leads to error accumulation after pseudo-label-guided enhancement. In this paper, we propose a novel Partial label-based Self-training framework (PaSta for short) that leverages partial label learning technique to overcome the limitations of existing methods. Specifically, PaSta first trains multiple annotators to comprehensively capture the class distribution of nodes and aggregates their predictions to construct high-quality partial labels. Subsequently, we design a partial label-based classification model with two well-crafted loss functions to guide the model learning at both label and representation spaces. To further enhance the robustness against noisy labels, we introduce a self-training strategy where the labels refined by partial label learning are then used to further optimize the annotators in a closed-loop iterative manner. Extensive experiments on five datasets demonstrate that, compared with existing state-of-the-art methods, PaSta achieves an average improvement of 1.1% in classification performance under various noise settings.