Cervical cancer remains a major worldwide health issue, with early identification and risk assessment playing critical roles in effective preventive interventions. This paper presents the Cervix-AID-Net model for cervical precancer risk classification. The study designs and evaluates the proposed Cervix-AID-Net model based on patients colposcopy images. The model comprises a Convolutional Block Attention Module (CBAM) and convolutional layers that extract interpretable and representative features of colposcopic images to distinguish high-risk and low-risk cervical precancer. In addition, the proposed Cervix-AID-Net model integrates four explainable techniques, namely gradient class activation maps, Local Interpretable Model-agnostic Explanations, CartoonX, and pixel rate distortion explanation based on output feature maps and input features. The evaluation using holdout and ten-fold cross-validation techniques yielded a classification accuracy of 99.33\% and 99.81\%. The analysis revealed that CartoonX provides meticulous explanations for the decision of the Cervix-AID-Net model due to its ability to provide the relevant piece-wise smooth part of the image. The effect of Gaussian noise and blur on the input shows that the performance remains unchanged up to Gaussian noise of 3\% and blur of 10\%, while the performance reduces thereafter. A comparison study of the proposed model's performance compared to other deep learning approaches highlights the Cervix-AID-Net model's potential as a supplemental tool for increasing the effectiveness of cervical precancer risk assessment. The proposed method, which incorporates the CBAM and explainable artificial integration, has the potential to influence cervical cancer prevention and early detection, improving patient outcomes and lowering the worldwide burden of this preventable disease.

Wildfires pose a significantly increasing hazard to global ecosystems due to the climate crisis. Due to its complex nature, there is an urgent need for innovative approaches to wildfire prediction, such as machine learning. This research took a unique approach, differentiating from classical supervised learning, and addressed the gap in unsupervised wildfire prediction using autoencoders and clustering techniques for anomaly detection. Historical weather and normalised difference vegetation index datasets of Australia for 2005 - 2021 were utilised. Two main unsupervised approaches were analysed. The first used a deep autoencoder to obtain latent features, which were then fed into clustering models, isolation forest, local outlier factor and one-class SVM for anomaly detection. The second approach used a deep autoencoder to reconstruct the input data and use reconstruction errors to identify anomalies. Long Short-Term Memory (LSTM) autoencoders and fully connected (FC) autoencoders were employed in this part, both in an unsupervised way learning only from nominal data. The FC autoencoder outperformed its counterparts, achieving an accuracy of 0.71, an F1-score of 0.74, and an MCC of 0.42. These findings highlight the practicality of this method, as it effectively predicts wildfires in the absence of ground truth, utilising an unsupervised learning technique.

NASA's forthcoming Lunar Gateway space station, which will be uncrewed most of the time, will need to operate with an unprecedented level of autonomy. Enhancing autonomy on the Gateway presents several unique challenges, one of which is to equip the Canadarm3, the Gateway's external robotic system, with the capability to perform worksite monitoring. Monitoring will involve using the arm's inspection cameras to detect any anomalies within the operating environment, a task complicated by the widely-varying lighting conditions in space. In this paper, we introduce the visual anomaly detection and localization task for space applications and establish a benchmark with our novel synthetic dataset called ALLO (for Anomaly Localization in Lunar Orbit). We develop a complete data generation pipeline to create ALLO, which we use to evaluate the performance of state-of-the-art visual anomaly detection algorithms. Given the low tolerance for risk during space operations and the lack of relevant data, we emphasize the need for novel, robust, and accurate anomaly detection methods to handle the challenging visual conditions found in lunar orbit and beyond.

While deep learning has revolutionized computer-aided drug discovery, the AI community has predominantly focused on model innovation and placed less emphasis on establishing best benchmarking practices. We posit that without a sound model evaluation framework, the AI community's efforts cannot reach their full potential, thereby slowing the progress and transfer of innovation into real-world drug discovery. Thus, in this paper, we seek to establish a new gold standard for small molecule drug discovery benchmarking, WelQrate. Specifically, our contributions are threefold: WelQrate Dataset Collection - we introduce a meticulously curated collection of 9 datasets spanning 5 therapeutic target classes. Our hierarchical curation pipelines, designed by drug discovery experts, go beyond the primary high-throughput screen by leveraging additional confirmatory and counter screens along with rigorous domain-driven preprocessing, such as Pan-Assay Interference Compounds (PAINS) filtering, to ensure the high-quality data in the datasets; WelQrate Evaluation Framework - we propose a standardized model evaluation framework considering high-quality datasets, featurization, 3D conformation generation, evaluation metrics, and data splits, which provides a reliable benchmarking for drug discovery experts conducting real-world virtual screening; Benchmarking - we evaluate model performance through various research questions using the WelQrate dataset collection, exploring the effects of different models, dataset quality, featurization methods, and data splitting strategies on the results. In summary, we recommend adopting our proposed WelQrate as the gold standard in small molecule drug discovery benchmarking. The WelQrate dataset collection, along with the curation codes, and experimental scripts are all publicly available at WelQrate.org.

Disaggregated evaluation -- estimation of performance of a machine learning model on different subpopulations -- is a core task when assessing performance and group-fairness of AI systems. A key challenge is that evaluation data is scarce, and subpopulations arising from intersections of attributes (e.g., race, sex, age) are often tiny. Today, it is common for multiple clients to procure the same AI model from a model developer, and the task of disaggregated evaluation is faced by each customer individually. This gives rise to what we call the multi-task disaggregated evaluation problem, wherein multiple clients seek to conduct a disaggregated evaluation of a given model in their own data setting (task). In this work we develop a disaggregated evaluation method called SureMap that has high estimation accuracy for both multi-task and single-task disaggregated evaluations of blackbox models. SureMap's efficiency gains come from (1) transforming the problem into structured simultaneous Gaussian mean estimation and (2) incorporating external data, e.g., from the AI system creator or from their other clients. Our method combines maximum a posteriori (MAP) estimation using a well-chosen prior together with cross-validation-free tuning via Stein's unbiased risk estimate (SURE). We evaluate SureMap on disaggregated evaluation tasks in multiple domains, observing significant accuracy improvements over several strong competitors.

Predicting stroke risk is a complex challenge that can be enhanced by integrating diverse clinically available data modalities. This study introduces a self-supervised multimodal framework that combines 3D brain imaging, clinical data, and image-derived features to improve stroke risk prediction prior to onset. By leveraging large unannotated clinical datasets, the framework captures complementary and synergistic information across image and tabular data modalities. Our approach is based on a contrastive learning framework that couples contrastive language-image pretraining with an image-tabular matching module, to better align multimodal data representations in a shared latent space. The model is trained on the UK Biobank, which includes structural brain MRI and clinical data. We benchmark its performance against state-of-the-art unimodal and multimodal methods using tabular, image, and image-tabular combinations under diverse frozen and trainable model settings. The proposed model outperformed self-supervised tabular (image) methods by 2.6% (2.6%) in ROC-AUC and by 3.3% (5.6%) in balanced accuracy. Additionally, it showed a 7.6% increase in balanced accuracy compared to the best multimodal supervised model. Through interpretable tools, our approach demonstrated better integration of tabular and image data, providing richer and more aligned embeddings. Gradient-weighted Class Activation Mapping heatmaps further revealed activated brain regions commonly associated in the literature with brain aging, stroke risk, and clinical outcomes. This robust self-supervised multimodal framework surpasses state-of-the-art methods for stroke risk prediction and offers a strong foundation for future studies integrating diverse data modalities to advance clinical predictive modelling.

Visual prompting (VP) is a new technique that adapts well-trained frozen models for source domain tasks to target domain tasks. This study examines VP's benefits for black-box model-level backdoor detection. The visual prompt in VP maps class subspaces between source and target domains. We identify a misalignment, termed class subspace inconsistency, between clean and poisoned datasets. Deep neural networks (DNNs) are commonly used in complex applications but require extensive computational power, leading to significant costs. However, DNNs can include backdoors (Gu et al., 2017; Liu et al., 2018b; Tang et al., 2021; Qi et al., 2023b; Nguyen & Tran, 2021; Chen et al., 2017), which manipulate model responses to inputs with specific triggers (like certain pixel patterns) while functioning correctly on other inputs. In backdoor attacks, attackers embed these triggers in the training data, leading the model to associate the trigger with a particular outcome and misclassify inputs containing it. Black-box backdoor detection, which uses only blackbox queries to the suspicious model (i.e., the model to be inspected), is gaining attention.

In this paper, we address the critical need for interpretable and uncertainty-aware machine learning models in the context of online learning for high-risk industries, particularly cyber-security. While deep learning and other complex models have demonstrated impressive predictive capabilities, their opacity and lack of uncertainty quantification present significant questions about their trustworthiness. We propose a novel pipeline for online supervised learning problems in cyber-security, that harnesses the inherent interpretability and uncertainty awareness of Additive Gaussian Processes (AGPs) models. Our approach aims to balance predictive performance with transparency while improving the scalability of AGPs, which represents their main drawback, potentially enabling security analysts to better validate threat detection, troubleshoot and reduce false positives, and generally make trustworthy, informed decisions. This work contributes to the growing field of interpretable AI by proposing a class of models that can be significantly beneficial for high-stake decision problems such as the ones typical of the cyber-security domain. The source code is available.

Embedding-as-a-Service (EaaS) has emerged as a successful business pattern but faces significant challenges related to various forms of copyright infringement, including API misuse and different attacks. Various studies have proposed backdoor-based watermarking schemes to protect the copyright of EaaS services. In this paper, we reveal that previous watermarking schemes possess semantic-independent characteristics and propose the Semantic Perturbation Attack (SPA). Our theoretical and experimental analyses demonstrate that this semantic-independent nature makes current watermarking schemes vulnerable to adaptive attacks that exploit semantic perturbations test to bypass watermark verification. To address this vulnerability, we propose the Semantic Aware Watermarking (SAW) scheme, a robust defense mechanism designed to resist SPA, by injecting a watermark that adapts to the text semantics. Extensive experimental results across multiple datasets demonstrate that the True Positive Rate (TPR) for detecting watermarked samples under SPA can reach up to more than 95%, rendering previous watermarks ineffective. Meanwhile, our watermarking scheme can resist such attack while ensuring the watermark verification capability. Our code is available at https://github.com/Zk4-ps/EaaS-Embedding-Watermark.

Protein language models (PLMs) have demonstrated remarkable success in protein modeling and design, yet their internal mechanisms for predicting structure and function remain poorly understood. Here we present a systematic approach to extract and analyze interpretable features from PLMs using sparse autoencoders (SAEs). By training SAEs on embeddings from the PLM ESM-2, we identify up to 2,548 human-interpretable latent features per layer that strongly correlate with up to 143 known biological concepts such as binding sites, structural motifs, and functional domains. In contrast, examining individual neurons in ESM-2 reveals up to 46 neurons per layer with clear conceptual alignment across 15 known concepts, suggesting that PLMs represent most concepts in superposition. Beyond capturing known annotations, we show that ESM-2 learns coherent concepts that do not map onto existing annotations and propose a pipeline using language models to automatically interpret novel latent features learned by the SAEs. As practical applications, we demonstrate how these latent features can fill in missing annotations in protein databases and enable targeted steering of protein sequence generation. Our results demonstrate that PLMs encode rich, interpretable representations of protein biology and we propose a systematic framework to extract and analyze these latent features. In the process, we recover both known biology and potentially new protein motifs. As community resources, we introduce InterPLM (interPLM.ai), an interactive visualization platform for exploring and analyzing learned PLM features, and release code for training and analysis at github.com/ElanaPearl/interPLM.