While significant progress has been made in conventional fairness-aware machine learning (ML) and differentially private ML (DPML), the fairness of privacy protection across groups remains underexplored. Existing studies have proposed methods to assess group privacy risks, but these are based on the average-case privacy risks of data records. Such approaches may underestimate the group privacy risks, thereby potentially underestimating the disparity across group privacy risks. Moreover, the current method for assessing the worst-case privacy risks of data records is time-consuming, limiting their practical applicability. To address these limitations, we introduce a novel membership inference game that can efficiently audit the approximate worst-case privacy risks of data records. Experimental results demonstrate that our method provides a more stringent measurement of group privacy risks, yielding a reliable assessment of the disparity in group privacy risks. Furthermore, to promote privacy protection fairness in DPML, we enhance the standard DP-SGD algorithm with an adaptive group-specific gradient clipping strategy, inspired by the design of canaries in differential privacy auditing studies. Extensive experiments confirm that our algorithm effectively reduces the disparity in group privacy risks, thereby enhancing the fairness of privacy protection in DPML.

The goal of coreset selection methods is to identify representative subsets of datasets for efficient model training. Yet, existing methods often ignore the possibility of annotation errors and require fixed pruning ratios, making them impractical in real-world settings. We present HyperCore, a robust and adaptive coreset selection framework designed explicitly for noisy environments. HyperCore leverages lightweight hypersphere models learned per class, embedding in-class samples close to a hypersphere center while naturally segregating out-of-class samples based on their distance. By using Youden's J statistic, HyperCore can adaptively select pruning thresholds, enabling automatic, noise-aware data pruning without hyperparameter tuning. Our experiments reveal that HyperCore consistently surpasses state-of-the-art coreset selection methods, especially under noisy and low-data regimes. HyperCore effectively discards mislabeled and ambiguous points, yielding compact yet highly informative subsets suitable for scalable and noise-free learning.

The broad capabilities and substantial resources required to train Large Language Models (LLMs) make them valuable intellectual property, yet they remain vulnerable to copyright infringement, such as unauthorized use and model theft. LLM fingerprinting, a non-intrusive technique that compares the distinctive features (i.e., fingerprint) of LLMs to identify whether an LLM is derived from another, offers a promising solution to copyright auditing. However, its reliability remains uncertain due to the prevalence of diverse model modifications and the lack of standardized evaluation. In this SoK, we present the first comprehensive study of the emerging LLM fingerprinting. We introduce a unified framework and taxonomy that structures the field: white-box methods are classified based on their feature source as static, forward-pass, or backward-pass fingerprinting, while black-box methods are distinguished by their query strategy as either untargeted or targeted. Furthermore, we propose LeaFBench, the first systematic benchmark for evaluating LLM fingerprinting under realistic deployment scenarios. Built upon 7 mainstream foundation models and comprising 149 distinct model instances, LeaFBench integrates 13 representative post-development techniques, spanning both parameter-altering methods (e.g., fine-tuning, quantization) and parameter-independent techniques (e.g., system prompts, RAG). Extensive experiments on LeaFBench reveal the strengths and weaknesses of existing methods, thereby outlining future research directions and critical open problems in this emerging field. The code is available at https://github.com/shaoshuo-ss/LeaFBench.

LLM-based solvers have emerged as a promising means of automating problem modeling and solving. However, they remain unreliable and often depend on iterative repair loops that result in significant latency. We introduce OptiHive, a framework that enhances any solver-generation pipeline to produce higher-quality solvers from natural-language descriptions of optimization problems. OptiHive uses a single batched generation to produce diverse components (solvers, problem instances, and validation tests) and filters out erroneous components to ensure fully interpretable outputs. Accounting for the imperfection of the generated components, we employ a statistical model to infer their true performance, enabling principled uncertainty quantification and solver selection. On tasks ranging from traditional optimization problems to challenging variants of the Multi-Depot Vehicle Routing Problem, OptiHive significantly outperforms baselines, increasing the optimality rate from 5% to 92% on the most complex problems.

Automated interpretation of CT images--particularly localizing and describing abnormal findings across multi-plane and whole-body scans--remains a significant challenge in clinical radiology. This work aims to address this challenge through four key contributions: (i) On taxonomy, we collaborate with senior radiologists to propose a comprehensive hierarchical classification system, with 404 representative abnormal findings across all body regions; (ii) On data, we contribute a dataset containing over 14.5K CT images from multiple planes and all human body regions, and meticulously provide grounding annotations for over 19K abnormalities, each linked to the detailed description and cast into the taxonomy; (iii) On model development, we propose OmniAbnorm-CT, which can automatically ground and describe abnormal findings on multi-plane and whole-body CT images based on text queries, while also allowing flexible interaction through visual prompts; (iv) On evaluation, we establish three representative tasks based on real clinical scenarios, and introduce a clinically grounded metric to assess abnormality descriptions. Through extensive experiments, we show that OmniAbnorm-CT can significantly outperform existing methods in both internal and external validations, and across all the tasks.

In the landscape of Fact-based Judgment Prediction and Explanation (FJPE), reliance on factual data is essential for developing robust and realistic AI-driven decision-making tools. This paper introduces TathyaNyaya, the largest annotated dataset for FJPE tailored to the Indian legal context, encompassing judgments from the Supreme Court of India and various High Courts. Derived from the Hindi terms "Tathya" (fact) and "Nyaya" (justice), the TathyaNyaya dataset is uniquely designed to focus on factual statements rather than complete legal texts, reflecting real-world judicial processes where factual data drives outcomes. Complementing this dataset, we present FactLegalLlama, an instruction-tuned variant of the LLaMa-3-8B Large Language Model (LLM), optimized for generating high-quality explanations in FJPE tasks. Finetuned on the factual data in TathyaNyaya, FactLegalLlama integrates predictive accuracy with coherent, contextually relevant explanations, addressing the critical need for transparency and interpretability in AI-assisted legal systems. Our methodology combines transformers for binary judgment prediction with FactLegalLlama for explanation generation, creating a robust framework for advancing FJPE in the Indian legal domain. TathyaNyaya not only surpasses existing datasets in scale and diversity but also establishes a benchmark for building explainable AI systems in legal analysis. The findings underscore the importance of factual precision and domain-specific tuning in enhancing predictive performance and interpretability, positioning TathyaNyaya and FactLegalLlama as foundational resources for AI-assisted legal decision-making.

This paper presents a fairness-audited and interpretable machine learning framework for predicting polycystic ovary syndrome (PCOS), designed to evaluate model performance and identify diagnostic disparities across patient subgroups. The framework integrated SHAP-based feature attributions with demographic audits to connect predictive explanations with observed disparities for actionable insights. Probabilistic calibration metrics (Brier Score and Expected Calibration Error) are incorporated to ensure reliable risk predictions across subgroups. Random Forest, SVM, and XGBoost models were trained with isotonic and Platt scaling for calibration and fairness comparison. A calibrated Random Forest achieved a high predictive accuracy of 90.8%. SHAP analysis identified follicle count, weight gain, and menstrual irregularity as the most influential features, which are consistent with the Rotterdam diagnostic criteria. Although the SVM with isotonic calibration achieved the lowest calibration error (ECE = 0.0541), the Random Forest model provided a better balance between calibration and interpretability (Brier = 0.0678, ECE = 0.0666). Therefore, it was selected for detailed fairness and SHAP analyses. Subgroup analysis revealed that the model performed best among women aged 25-35 (accuracy 90.9%) but underperformed in those under 25 (69.2%), highlighting age-related disparities. The model achieved perfect precision in obese women and maintained high recall in lean PCOS cases, demonstrating robustness across phenotypes. Finally, a Streamlit-based web interface enables real-time PCOS risk assessment, Rotterdam criteria evaluation, and interactive 'what-if' analysis, bridging the gap between AI research and clinical usability.

Graph neural networks (GNNs) are frequently used for knowledge graph completion. Their black-box nature has motivated work that uses sound logical rules to explain predictions and characterise their expressivity. However, despite the prevalence of GNNs that use mean as an aggregation function, explainability and expressivity results are lacking for them. We consider GNNs with mean aggregation and non-negative weights (MAGNNs), proving the precise class of monotonic rules that can be sound for them, as well as providing a restricted fragment of first-order logic to explain any MAGNN prediction. Our experiments show that restricting mean-aggregation GNNs to have non-negative weights yields comparable or improved performance on standard inductive benchmarks, that sound rules are obtained in practice, that insightful explanations can be generated in practice, and that the sound rules can expose issues in the trained models.

Machine learning algorithms are increasingly deployed in critical domains such as finance, healthcare, and criminal justice [1]. The increasing popularity of algorithmic decision-making has stimulated interest in algorithmic fairness within the academic community. Researchers have introduced various fairness definitions that quantify disparities between privileged and protected groups, use causal inference to determine the impact of race on model predictions, and that test calibration of probability predictions from the model. Existing literature does not provide a way in which to assess whether observed disparities between groups are statistically significant or merely due to chance. This paper introduces a rigorous framework for testing the statistical significance of fairness violations by leveraging k-fold cross-validation [2] to generate sampling distributions of fairness metrics. This paper introduces statistical tests that can be used to identify statistically significant violations of fairness metrics based on disparities between predicted and actual outcomes, model calibration, and causal inference techniques [1]. We demonstrate this approach by testing recidivism forecasting algorithms trained on data from the National Institute of Justice. Our findings reveal that machine learning algorithms used for recidivism forecasting exhibit statistically significant bias against Black individuals under several fairness definitions, while also exhibiting no bias or bias against White individuals under other definitions. The results from this paper underscore the importance of rigorous and robust statistical testing while evaluating algorithmic decision-making systems.

The transformer architecture has demonstrated strong performance in classification tasks involving structured and high-dimensional data. However, its success often hinges on large- scale training data and careful regularization to prevent overfitting. In this paper, we intro- duce a novel likelihood-guided variational Ising-based regularization framework for Vision Transformers (ViTs), which simultaneously enhances model generalization and dynamically prunes redundant parameters. The proposed variational Ising-based regularization approach leverages Bayesian sparsification techniques to impose structured sparsity on model weights, allowing for adaptive architecture search during training. Unlike traditional dropout-based methods, which enforce fixed sparsity patterns, the variational Ising-based regularization method learns task-adaptive regularization, improving both efficiency and interpretability. We evaluate our approach on benchmark vision datasets, including MNIST, Fashion-MNIST, CIFAR-10, and CIFAR-100, demonstrating improved generalization under sparse, complex data and allowing for principled uncertainty quantification on both weights and selection parameters. Additionally, we show that the Ising regularizer leads to better-calibrated probability estimates and structured feature selection through uncertainty-aware attention mechanisms. Our results highlight the effectiveness of structured Bayesian sparsification in enhancing transformer-based architectures, offering a principled alternative to standard regularization techniques.