class-imbalance

The paper introduces a controlled evaluation framework separating unique class support from representation allocation to determine when encoder adaptation remains worthwhile after classifier correction in long-tailed recognition, showing benchmark construction alone can flip this training decision.

The paper proposes a comprehensive data generation strategy combining traditional augmentation, CutMix, MixUp, and DCGAN-generated images to address data insufficiency and class imbalance for olive tree disease classification, evaluated using EfficientNetV2B0 and showing notable accuracy improvements.

提出 PAI-KNN，一种通过推理时先验调整而非内存修改来适应概念漂移与类不平衡的 KNN 数据流分类器，在合成数据流基准上取得了与最先进方法竞争的性能。

该论文提出一个理论框架，分析监督对比学习目标在长尾分布下是否能形成正则单纯形几何，并提出一种完全平衡的对比目标，可证明在最优时形成正则单纯形，从而在长尾识别和迁移学习任务上取得更好的性能。

RGH-GNN 提出一种风险门控的异构图神经网络，将 LightGBM 生成的交易级风险分数作为边权重引导消息传递，以提升金融欺诈检测性能，在 IEEE-CIS 数据集上取得 0.914 的 AUC-ROC 并超越所有基线。

CRISP is a scalable coreset method for imbalanced tabular learning that efficiently reduces dataset size while maintaining high accuracy in tasks like fraud detection.

This paper presents the first comprehensive empirical study on multi-task learning for predictive process monitoring, showing improvements over single-task learning in next-activity prediction and class imbalance mitigation.

The paper presents a nine-voter ensemble system using error-independent LLMs for harmful content detection in German social media, achieving first place in GermEval 2026 shared task across four subtasks by addressing class imbalance.

RoBell-RVFL is a robust randomized neural network that uses a generalized bell membership function to handle class imbalance and noisy data, achieving superior performance on benchmark datasets.

This study investigates how class imbalance and batch processing affect LLM-based screening in systematic reviews, revealing that batch processing significantly alters decision-making behavior while prevalence metadata has minimal impact.

Introduces CALCoDe, a post-hoc reliability layer for frozen medical vision-language models that mitigates class-tail undercoverage under clinical shift, achieving strong worst-class accepted coverage across multiple dermatology shifts and VLM backbones.

Introduces Sharpness-Guided Equilibrium Sampling (SGS) that dynamically adjusts sampling probabilities using sharpness estimates to achieve balanced flat minima in long-tailed learning, achieving significant gains on CIFAR-100 LT and ImageNet-LT.

A federated learning research project reveals that global accuracy can mask catastrophic failure on minority attack classes in network intrusion detection, showing that per-client performance and aggregation method choice are critical for rare attack detection.

This paper compares cascaded and joint multi-task modeling for hierarchical offensive language detection, finding that cascaded architectures achieve higher accuracy at the cost of increased parameters and inference latency, and that class-imbalance handling strategies should be verified via ablation.

Proposes a Hierarchy-Aware RoBERTa framework for classifying cybersecurity vulnerabilities in the CWE taxonomy, demonstrating that hierarchy-aware representation learning is more effective than oversampling techniques for handling class imbalance.

This paper reveals that standard marginal conformal prediction fails to cover minority classes in imbalanced virtual screening datasets, and demonstrates that class-conditional (Mondrian) conformal prediction restores per-class coverage.

The paper proposes FedCGNM, a client-side optimizer that groups classes and normalizes momentum per group to address class imbalance in federated learning, along with FedHOO for efficient hyperparameter exploration. Empirical results show consistent improvements over baselines.

This paper introduces SemiScope, an analysis tool designed to disentangle the effects of classifier tuning from joint SSL and classifier optimization in semi-supervised security classification. Results show that most performance gains attributed to joint optimization can be recovered by simply tuning the classifier and its decision threshold with Bayesian optimization.

Proposes BP-TTA, a test-time adaptation method that handles both class imbalance and continual domain shifts by combining batch-balanced sampling with prototype-guided constraints, achieving state-of-the-art performance in dynamic streaming scenarios.

This paper investigates how training dynamics of neural networks for software defect prediction are affected by coupled data-quality issues such as class imbalance and overlap, proposing an interaction-aware empirical protocol.