Large language models (LLMs) are increasingly utilized in applications where system prompts, which guide model outputs, play a crucial role. These prompts often contain business logic and sensitive information, making their protection essential. However, adversarial and even regular user queries can exploit LLM vulnerabilities to expose these hidden prompts. To address this issue, we propose PromptKeeper, a robust defense mechanism designed to safeguard system prompts. PromptKeeper tackles two core challenges: reliably detecting prompt leakage and mitigating side-channel vulnerabilities when leakage occurs. By framing detection as a hypothesis-testing problem, PromptKeeper effectively identifies both explicit and subtle leakage. Upon detection, it regenerates responses using a dummy prompt, ensuring that outputs remain indistinguishable from typical interactions when no leakage is present. PromptKeeper ensures robust protection against prompt extraction attacks via either adversarial or regular queries, while preserving conversational capability and runtime efficiency during benign user interactions.

We consider the problem of learning convolution operators associated to compact Abelian groups. We study a regularization-based approach and provide corresponding learning guarantees, discussing natural regularity condition on the convolution kernel. More precisely, we assume the convolution kernel is a function in a translation invariant Hilbert space and analyze a natural ridge regression (RR) estimator. Building on existing results for RR, we characterize the accuracy of the estimator in terms of finite sample bounds. Interestingly, regularity assumptions which are classical in the analysis of RR, have a novel and natural interpretation in terms of space/frequency localization. Theoretical results are illustrated by numerical simulations.

Proposed in Hyv\"arinen (2005), score matching is a parameter estimation procedure that does not require computation of distributional normalizing constants. In this work we utilize the geometric median of means to develop a robust score matching procedure that yields consistent parameter estimates in settings where the observed data has been contaminated. A special appeal of the proposed method is that it retains convexity in exponential family models. The new method is therefore particularly attractive for non-Gaussian, exponential family graphical models where evaluation of normalizing constants is intractable. Support recovery guarantees for such models when contamination is present are provided. Additionally, support recovery is studied in numerical experiments and on a precipitation dataset. We demonstrate that the proposed robust score matching estimator performs comparably to the standard score matching estimator when no contamination is present but greatly outperforms this estimator in a setting with contamination.

While text-to-image models offer numerous benefits, they also pose significant societal risks. Detecting AI-generated images is crucial for mitigating these risks. Detection methods can be broadly categorized into passive and watermark-based approaches: passive detectors rely on artifacts present in AI-generated images, whereas watermark-based detectors proactively embed watermarks into such images. A key question is which type of detector performs better in terms of effectiveness, robustness, and efficiency. However, the current literature lacks a comprehensive understanding of this issue. In this work, we aim to bridge that gap by developing ImageDetectBench, the first comprehensive benchmark to compare the effectiveness, robustness, and efficiency of passive and watermark-based detectors. Our benchmark includes four datasets, each containing a mix of AI-generated and non-AI-generated images. We evaluate five passive detectors and four watermark-based detectors against eight types of common perturbations and three types of adversarial perturbations. Our benchmark results reveal several interesting findings. For instance, watermark-based detectors consistently outperform passive detectors, both in the presence and absence of perturbations. Based on these insights, we provide recommendations for detecting AI-generated images, e.g., when both types of detectors are applicable, watermark-based detectors should be the preferred choice. Our code and data are publicly available at https://github.com/moyangkuo/ImageDetectBench.git.

This paper introduces two novel, outlyingness scores (OSs) based on Cluster Catch Digraphs (CCDs): Outbound Outlyingness Score (OOS) and Inbound Outlyingness Score (IOS). These scores enhance the interpretability of outlier detection results. Both OSs employ graph-, density-, and distribution-based techniques, tailored to high-dimensional data with varying cluster shapes and intensities. OOS evaluates the outlyingness of a point relative to its nearest neighbors, while IOS assesses the total ``influence" a point receives from others within its cluster. Both OSs effectively identify global and local outliers, invariant to data collinearity. Moreover, IOS is robust to the masking problems. With extensive Monte Carlo simulations, we compare the performance of both OSs with CCD-based, traditional, and state-of-the-art outlier detection methods. Both OSs exhibit substantial overall improvements over the CCD-based methods in both artificial and real-world data sets, particularly with IOS, which delivers the best overall performance among all the methods, especially in high-dimensional settings. Keywords: Outlier detection, Outlyingness score, Graph-based clustering, Cluster catch digraphs, High-dimensional data.

Sparse autoencoders (SAEs) have gained a lot of attention as a promising tool to improve the interpretability of large language models (LLMs) by mapping the complex superposition of polysemantic neurons into monosemantic features and composing a sparse dictionary of words. However, traditional performance metrics like Mean Squared Error and L0 sparsity ignore the evaluation of the semantic representational power of SAEs -- whether they can acquire interpretable monosemantic features while preserving the semantic relationship of words. For instance, it is not obvious whether a learned sparse feature could distinguish different meanings in one word. In this paper, we propose a suite of evaluations for SAEs to analyze the quality of monosemantic features by focusing on polysemous words. Our findings reveal that SAEs developed to improve the MSE-L0 Pareto frontier may confuse interpretability, which does not necessarily enhance the extraction of monosemantic features. The analysis of SAEs with polysemous words can also figure out the internal mechanism of LLMs; deeper layers and the Attention module contribute to distinguishing polysemy in a word. Our semantics focused evaluation offers new insights into the polysemy and the existing SAE objective and contributes to the development of more practical SAEs.

The inaccurate translation of numbers can lead to significant security issues, ranging from financial setbacks to medical inaccuracies. While large language models (LLMs) have made significant advancements in machine translation, their capacity for translating numbers has not been thoroughly explored. This study focuses on evaluating the reliability of LLM-based machine translation systems when handling numerical data. In order to systematically test the numerical translation capabilities of currently open source LLMs, we have constructed a numerical translation dataset between Chinese and English based on real business data, encompassing ten types of numerical translation. Experiments on the dataset indicate that errors in numerical translation are a common issue, with most open-source LLMs faltering when faced with our test scenarios. Especially when it comes to numerical types involving large units like ``million", ``billion", and "yi", even the latest llama3.1 8b model can have error rates as high as 20%. Finally, we introduce three potential strategies to mitigate the numerical mistranslations for large units.

While machine learning offers diverse techniques suitable for exploring various medical research questions, a cohesive synergistic framework can facilitate the integration and understanding of new approaches within unified model development and interpretation. We therefore introduce the Medical Artificial Intelligence Toolbox (MAIT), an explainable, open-source Python pipeline for developing and evaluating binary classification, regression, and survival models on tabular datasets. MAIT addresses key challenges (e.g., high dimensionality, class imbalance, mixed variable types, and missingness) while promoting transparency in reporting (TRIPOD+AI compliant). Offering automated configurations for beginners and customizable source code for experts, MAIT streamlines two primary use cases: Discovery (feature importance via unified scoring, e.g., SHapley Additive exPlanations - SHAP) and Prediction (model development and deployment with optimized solutions). Moreover, MAIT proposes new techniques including fine-tuning of probability threshold in binary classification, translation of cumulative hazard curves to binary classification, enhanced visualizations for model interpretation for mixed data types, and handling censoring through semi-supervised learning, to adapt to a wide set of data constraints and study designs. We provide detailed tutorials on GitHub, using four open-access data sets, to demonstrate how MAIT can be used to improve implementation and interpretation of ML models in medical research.

Congenital heart disease is among the most common fetal abnormalities and birth defects. Despite identifying numerous risk factors influencing its onset, a comprehensive understanding of its genesis and management across diverse populations remains limited. Recent advancements in machine learning have demonstrated the potential for leveraging patient data to enable early congenital heart disease detection. Over the past seven years, researchers have proposed various data-driven and algorithmic solutions to address this challenge. This paper presents a systematic review of congential heart disease recognition using machine learning, conducting a meta-analysis of 432 references from leading journals published between 2018 and 2024. A detailed investigation of 74 scholarly works highlights key factors, including databases, algorithms, applications, and solutions. Additionally, the survey outlines reported datasets used by machine learning experts for congenital heart disease recognition. Using a systematic literature review methodology, this study identifies critical challenges and opportunities in applying machine learning to congenital heart disease.

The Large Hadron Collider at CERN produces immense volumes of complex data from high-energy particle collisions, demanding sophisticated analytical techniques for effective interpretation. Neural Networks, including Graph Neural Networks, have shown promise in tasks such as event classification and object identification by representing collisions as graphs. However, while Graph Neural Networks excel in predictive accuracy, their "black box" nature often limits their interpretability, making it difficult to trust their decision-making processes. In this paper, we propose a novel approach that combines a Graph Transformer model with Mixture-of-Expert layers to achieve high predictive performance while embedding interpretability into the architecture. By leveraging attention maps and expert specialization, the model offers insights into its internal decision-making, linking predictions to physics-informed features. We evaluate the model on simulated events from the ATLAS experiment, focusing on distinguishing rare Supersymmetric signal events from Standard Model background. Our results highlight that the model achieves competitive classification accuracy while providing interpretable outputs that align with known physics, demonstrating its potential as a robust and transparent tool for high-energy physics data analysis. This approach underscores the importance of explainability in machine learning methods applied to high energy physics, offering a path toward greater trust in AI-driven discoveries.