Daily Papers

Metabolomics knowledge is distributed across heterogeneous resources and remains difficult to translate into predictive representations. We developed MetaboLLM, a metabolomics-specialized large language model adapted through continual pretraining, supervised fine-tuning, and structured retrieval, together with MetaboLLM-GIN, which converts generated biochemical descriptions into metabolite graphs for patient-level prediction using a graph isomorphism network. Across four backbone families, MetaboLLM outperformed corresponding base and medically adapted models on metabolomics knowledge, relational, and description tasks, and transferred to an external public benchmark. MetaboLLM-GIN achieved the highest AUC for stress hyperglycemia prediction after coronary artery bypass grafting (0.8616) and postmenopausal hormone-regimen classification (0.8123), outperforming conventional models, alternative graph constructions, and graphs generated from unadapted or non-retrieval LLM configurations. Model interpretation further produced biologically meaningful findings in both applications. These results show that domain-specialized language models can organize heterogeneous biochemical knowledge into predictive and interpretable metabolite graph representations.

Flashcard schedulers are tools that rely on 1) student models to predict the flashcards a student knows; and 2) teaching policies to schedule cards based on these predictions. Existing student models, however, only use flashcard-level features, like the student's past responses, ignoring the semantic ties of flashcards. Deep Knowledge Tracing (DKT) models can capture semantic relations with language models, but are inefficient, lack content-rich datasets for evaluation, and require robust teaching policies. To address these issues, we design KARL, a DKT-inspired student model that uses retrieval and BERT embeddings for efficient and accurate student recall predictions. To test KARL, we collect a new dataset of diverse study history on trivia questions. KARL bests existing student models in AUC and calibration error. Finally, we propose a novel teaching policy that exploits the predictive power of DKT models to deploy KARL online. Based on 27 learners and 32 6-day study trajectories, KARL shows the ability to enhance medium-term educational learning, proving its efficacy for scheduling.

Heart failure is a leading cause of global mortality, necessitating improved diagnostic strategies. Classical machine learning models struggle with challenges such as high-dimensional data, class imbalances, poor feature representations, and a lack of interpretability. While quantum machine learning holds promise, current hybrid models have not fully exploited quantum advantages. In this paper, we propose the Kolmogorov-Arnold Classical-Quantum Dual-Channel Neural Network (KACQ-DCNN), a novel hybrid architecture that replaces traditional multilayer perceptrons with Kolmogorov-Arnold Networks (KANs), enabling learnable univariate activation functions. Our KACQ-DCNN 4-qubit, 1-layer model outperforms 37 benchmark models, including 16 classical and 12 quantum neural networks, achieving an accuracy of 92.03%, with macro-average precision, recall, and F1 scores of 92.00%. It also achieved a ROC-AUC of 94.77%, surpassing other models by significant margins, as validated by paired t-tests with a significance threshold of 0.0056 (after Bonferroni correction). Ablation studies highlight the synergistic effect of classical-quantum integration, improving performance by about 2% over MLP variants. Additionally, LIME and SHAP explainability techniques enhance feature interpretability, while conformal prediction provides robust uncertainty quantification. Our results demonstrate that KACQ-DCNN improves cardiovascular diagnostics by combining high accuracy with interpretability and uncertainty quantification.

Deep AUC Maximization (DAM) is a new paradigm for learning a deep neural network by maximizing the AUC score of the model on a dataset. Most previous works of AUC maximization focus on the perspective of optimization by designing efficient stochastic algorithms, and studies on generalization performance of large-scale DAM on difficult tasks are missing. In this work, we aim to make DAM more practical for interesting real-world applications (e.g., medical image classification). First, we propose a new margin-based min-max surrogate loss function for the AUC score (named as AUC min-max-margin loss or simply AUC margin loss for short). It is more robust than the commonly used AUC square loss, while enjoying the same advantage in terms of large-scale stochastic optimization. Second, we conduct extensive empirical studies of our DAM method on four difficult medical image classification tasks, namely (i) classification of chest x-ray images for identifying many threatening diseases, (ii) classification of images of skin lesions for identifying melanoma, (iii) classification of mammogram for breast cancer screening, and (iv) classification of microscopic images for identifying tumor tissue. Our studies demonstrate that the proposed DAM method improves the performance of optimizing cross-entropy loss by a large margin, and also achieves better performance than optimizing the existing AUC square loss on these medical image classification tasks. Specifically, our DAM method has achieved the 1st place on Stanford CheXpert competition on Aug. 31, 2020. To the best of our knowledge, this is the first work that makes DAM succeed on large-scale medical image datasets. We also conduct extensive ablation studies to demonstrate the advantages of the new AUC margin loss over the AUC square loss on benchmark datasets. The proposed method is implemented in our open-sourced library LibAUC (www.libauc.org).

Selecting LLM-generated code candidates using LLM-generated tests is challenging because the tests themselves may be incorrect. Existing methods either treat all tests equally or rely on ad-hoc heuristics to filter unreliable tests. Yet determining test correctness requires knowing which codes are correct, creating a circular dependency. Our key insight is that we need not determine test correctness at all: test votes should rank, not merely count. What matters is not how many codes pass a test, but whether the test can distinguish correct from incorrect code. We break the circular dependency via leave-one-out evaluation: hold out one test, rank codes by their aggregate scores on all remaining tests, and measure whether the held-out test's pass/fail pattern agrees with this ranking. We formalize this agreement as the leave-one-out AUC~(LOO-AUC) and prove that the expected LOO-AUC is proportional to each test's ability to separate correct code from incorrect code. Building on this, we propose ACES~(AUC ConsistEncy Scoring) with two complementary variants: ACES-C provides closed-form weights that provably approximate the oracle in expectation under a mild assumption on average test quality; ACES-O drops this assumption and iteratively optimizes a differentiable LOO-AUC objective. Both operate solely on the binary pass matrix with negligible overhead, and achieve state-of-the-art Pass@k on multiple code generation benchmarks.

To maintain user trust, large language models (LLMs) should signal low confidence on examples where they are incorrect, instead of misleading the user. The standard approach of estimating confidence is to use the softmax probabilities of these models, but as of November 2023, state-of-the-art LLMs such as GPT-4 and Claude-v1.3 do not provide access to these probabilities. We first study eliciting confidence linguistically -- asking an LLM for its confidence in its answer -- which performs reasonably (80.5% AUC on GPT-4 averaged across 12 question-answering datasets -- 7% above a random baseline) but leaves room for improvement. We then explore using a surrogate confidence model -- using a model where we do have probabilities to evaluate the original model's confidence in a given question. Surprisingly, even though these probabilities come from a different and often weaker model, this method leads to higher AUC than linguistic confidences on 9 out of 12 datasets. Our best method composing linguistic confidences and surrogate model probabilities gives state-of-the-art confidence estimates on all 12 datasets (84.6% average AUC on GPT-4).

The contributions of model complexity, data volume, and feature modalities to knowledge graph-based drug repurposing remain poorly quantified under rigorous temporal validation. We constructed a pharmacology knowledge graph from ChEMBL 36 comprising 5,348 entities including 3,127 drugs, 1,156 proteins, and 1,065 indications. A strict temporal split was enforced with training data up to 2022 and testing data from 2023 to 2025, together with biologically verified hard negatives mined from failed assays and clinical trials. We benchmarked five knowledge graph embedding models and a standard graph neural network with 3.44 million parameters that incorporates drug chemical structure using a graph attention encoder and ESM-2 protein embeddings. Scaling experiments ranging from 0.78 to 9.75 million parameters and from 25 to 100 percent of the data, together with feature ablation studies, were used to isolate the contributions of model capacity, graph density, and node feature modalities. Removing the graph attention based drug structure encoder and retaining only topological embeddings combined with ESM-2 protein features improved drug protein PR-AUC from 0.5631 to 0.5785 while reducing VRAM usage from 5.30 GB to 353 MB. Replacing the drug encoder with Morgan fingerprints further degraded performance, indicating that explicit chemical structure representations can be detrimental for predicting pharmacological network interactions. Increasing model size beyond 2.44 million parameters yielded diminishing returns, whereas increasing training data consistently improved performance. External validation confirmed 6 of the top 14 novel predictions as established therapeutic indications. These results show that drug pharmacological behavior can be accurately predicted using target-centric information and drug network topology alone, without requiring explicit chemical structure representations.

The global outbreak of the Coronavirus 2019 (COVID-19) has overloaded worldwide healthcare systems. Computer-aided diagnosis for COVID-19 fast detection and patient triage is becoming critical. This paper proposes a novel self-knowledge distillation based self-supervised learning method for COVID-19 detection from chest X-ray images. Our method can use self-knowledge of images based on similarities of their visual features for self-supervised learning. Experimental results show that our method achieved an HM score of 0.988, an AUC of 0.999, and an accuracy of 0.957 on the largest open COVID-19 chest X-ray dataset.

We propose GE2E-KWS -- a generalized end-to-end training and evaluation framework for customized keyword spotting. Specifically, enrollment utterances are separated and grouped by keywords from the training batch and their embedding centroids are compared to all other test utterance embeddings to compute the loss. This simulates runtime enrollment and verification stages, and improves convergence stability and training speed by optimizing matrix operations compared to SOTA triplet loss approaches. To benchmark different models reliably, we propose an evaluation process that mimics the production environment and compute metrics that directly measure keyword matching accuracy. Trained with GE2E loss, our 419KB quantized conformer model beats a 7.5GB ASR encoder by 23.6% relative AUC, and beats a same size triplet loss model by 60.7% AUC. Our KWS models are natively streamable with low memory footprints, and designed to continuously run on-device with no retraining needed for new keywords (zero-shot).

Automated multi-label chest X-rays (CXR) image classification has achieved substantial progress in clinical diagnosis via utilizing sophisticated deep learning approaches. However, most deep models have high computational demands, which makes them less feasible for compact devices with low computational requirements. To overcome this problem, we propose a knowledge distillation (KD) strategy to create the compact deep learning model for the real-time multi-label CXR image classification. We study different alternatives of CNNs and Transforms as the teacher to distill the knowledge to a smaller student. Then, we employed explainable artificial intelligence (XAI) to provide the visual explanation for the model decision improved by the KD. Our results on three benchmark CXR datasets show that our KD strategy provides the improved performance on the compact student model, thus being the feasible choice for many limited hardware platforms. For instance, when using DenseNet161 as the teacher network, EEEA-Net-C2 achieved an AUC of 83.7%, 87.1%, and 88.7% on the ChestX-ray14, CheXpert, and PadChest datasets, respectively, with fewer parameters of 4.7 million and computational cost of 0.3 billion FLOPS.

Sparse keypoint extraction and matching underpin core tasks in geometric computer vision, including structure-from-motion, visual SLAM, augmented reality, and medical image registration. Learning robust local feature representations, however, typically requires accurate camera poses or depth supervision, which are often unavailable in real-world settings. Reinforcement learning (RL) has recently emerged as a promising alternative, requiring only the information if two images show the same scene or not. However, existing RL formulations such as RIPE rely on coarse binary rewards and carefully constructed negative training pairs, limiting training stability and descriptor discriminability. In this paper, we revisit RL-based keypoint learning and propose a reward that fully exploits the geometric consistency signal, deriving both reward and penalty from a single positive pair without contrasting against negatives. This richer signal provides sufficient supervisory contrast to learn discriminative detectors and descriptors from positive image pairs alone, enabling representation learning under extremely limited supervision. Furthermore, we show that the same RL objective can be extended to the matching stage by adapting LightGlue, raising AUC@5 on MegaDepth1500 from 56.58 to 59.65 and enabling weakly-supervised training of the full sparse matching pipeline from image pairs with partial visual overlap. We validate our approach on established benchmarks, demonstrating competitive results compared to fully-supervised methods. We further show that the method can be even trained on low texture medical video sequences, where camera poses are usually unavailable and standard SfM pipelines often fail. Code and data are available at https://github.com/fraunhoferhhi/RIPEpp .

If AI models can detect when they are being evaluated, the effectiveness of evaluations might be compromised. For example, models could have systematically different behavior during evaluations, leading to less reliable benchmarks for deployment and governance decisions. We investigate whether frontier language models can accurately classify transcripts based on whether they originate from evaluations or real-world deployment, a capability we call evaluation awareness. To achieve this, we construct a diverse benchmark of 1,000 prompts and transcripts from 61 distinct datasets. These span public benchmarks (e.g., MMLU, SWEBench), real-world deployment interactions, and agent trajectories from scaffolding frameworks (e.g., web-browsing agents). Frontier models clearly demonstrate above-random evaluation awareness (Gemini-2.5-Pro reaches an AUC of 0.83), but do not yet surpass our simple human baseline (AUC of 0.92). Furthermore, both AI models and humans are better at identifying evaluations in agentic settings compared to chat settings. Additionally, we test whether models can identify the purpose of the evaluation. Under multiple-choice and open-ended questioning, AI models far outperform random chance in identifying what an evaluation is testing for. Our results indicate that frontier models already exhibit a substantial, though not yet superhuman, level of evaluation-awareness. We recommend tracking this capability in future models.

Knitting patterns are a crucial component in the creation and design of knitted materials. Traditionally, these patterns were taught informally, but thanks to advancements in technology, anyone interested in knitting can use the patterns as a guide to start knitting. Perhaps because knitting is mostly a hobby, with the exception of industrial manufacturing utilising specialised knitting machines, the use of Al in knitting is less widespread than its application in other fields. However, it is important to determine whether knitted pattern classification using an automated system is viable. In order to recognise and classify knitting patterns. Using data augmentation and a transfer learning technique, this study proposes a deep learning model. The Inception ResNet-V2 is the main feature extraction and classification algorithm used in the model. Metrics like accuracy, logarithmic loss, F1-score, precision, and recall score were used to evaluate the model. The model evaluation's findings demonstrate high model accuracy, precision, recall, and F1 score. In addition, the AUC score for majority of the classes was in the range (0.7-0.9). A comparative analysis was done using other pretrained models and a ResNet-50 model with transfer learning and the proposed model evaluation results surpassed all others. The major limitation for this project is time, as with more time, there might have been better accuracy over a larger number of epochs.

Accurate prediction of knee osteoarthritis (KOA) progression from structural MRI has a potential to enhance disease understanding and support clinical trials. Prior art focused on manually designed imaging biomarkers, which may not fully exploit all disease-related information present in MRI scan. In contrast, our method learns relevant representations from raw data end-to-end using Deep Learning, and uses them for progression prediction. The method employs a 2D CNN to process the data slice-wise and aggregate the extracted features using a Transformer. Evaluated on a large cohort (n=4,866), the proposed method outperforms conventional 2D and 3D CNN-based models and achieves average precision of 0.58pm0.03 and ROC AUC of 0.78pm0.01. This paper sets a baseline on end-to-end KOA progression prediction from structural MRI. Our code is publicly available at https://github.com/MIPT-Oulu/OAProgressionMR.

The 7-point checklist (7PCL) is a widely used diagnostic tool in dermoscopy for identifying malignant melanoma by assigning point values to seven specific attributes. However, the traditional 7PCL is limited to distinguishing between malignant melanoma and melanocytic Nevi, and falls short in scenarios where multiple skin diseases with appearances similar to melanoma coexist. To address this limitation, we propose a novel diagnostic framework that integrates a clinical knowledge-based topological graph (CKTG) with a gradient diagnostic strategy featuring a data-driven weighting system (GD-DDW). The CKTG captures both the internal and external relationships among the 7PCL attributes, while the GD-DDW emulates dermatologists' diagnostic processes, prioritizing visual observation before making predictions. Additionally, we introduce a multimodal feature extraction approach leveraging a dual-attention mechanism to enhance feature extraction through cross-modal interaction and unimodal collaboration. This method incorporates meta-information to uncover interactions between clinical data and image features, ensuring more accurate and robust predictions. Our approach, evaluated on the EDRA dataset, achieved an average AUC of 88.6%, demonstrating superior performance in melanoma detection and feature prediction. This integrated system provides data-driven benchmarks for clinicians, significantly enhancing the precision of melanoma diagnosis.

Knowledge tracing (KT) refers to the problem of predicting future learner performance given their past performance in educational applications. Recent developments in KT using flexible deep neural network-based models excel at this task. However, these models often offer limited interpretability, thus making them insufficient for personalized learning, which requires using interpretable feedback and actionable recommendations to help learners achieve better learning outcomes. In this paper, we propose attentive knowledge tracing (AKT), which couples flexible attention-based neural network models with a series of novel, interpretable model components inspired by cognitive and psychometric models. AKT uses a novel monotonic attention mechanism that relates a learner's future responses to assessment questions to their past responses; attention weights are computed using exponential decay and a context-aware relative distance measure, in addition to the similarity between questions. Moreover, we use the Rasch model to regularize the concept and question embeddings; these embeddings are able to capture individual differences among questions on the same concept without using an excessive number of parameters. We conduct experiments on several real-world benchmark datasets and show that AKT outperforms existing KT methods (by up to 6% in AUC in some cases) on predicting future learner responses. We also conduct several case studies and show that AKT exhibits excellent interpretability and thus has potential for automated feedback and personalization in real-world educational settings.

Language models (LMs) should provide reliable confidence estimates to help users detect mistakes in their outputs and defer to human experts when necessary. Asking a language model to assess its confidence ("Score your confidence from 0-1.") is a natural way of evaluating its uncertainty. However, models struggle to provide absolute assessments of confidence (i.e. judging confidence in answering a question independent of other questions) and the coarse-grained scores they produce are not useful for evaluating the correctness of their answers. We propose relative confidence estimation, where we match up questions against each other and ask the model to make relative judgments of confidence ("Which question are you more confident in answering correctly?"). Treating each question as a "player" in a series of matchups against other questions and the model's preferences as match outcomes, we can use rank aggregation methods like Elo rating and Bradley-Terry to translate the model's confidence preferences into confidence scores. We evaluate relative confidence estimation against absolute confidence estimation and self-consistency confidence methods on five state-of-the-art LMs -- GPT-4, GPT-4o, Gemini 1.5 Pro, Claude 3.5 Sonnet, and Llama 3.1 405B -- across 14 challenging STEM, social science, and commonsense reasoning question answering tasks. Our results demonstrate that relative confidence estimation consistently provides more reliable confidence scores than absolute confidence estimation, with average gains of 3.5% in selective classification AUC over direct absolute confidence estimation methods and 1.7% over self-consistency approaches across all models and datasets.

Electrocardiograms (ECGs) are non-invasive diagnostic tools crucial for detecting cardiac arrhythmic diseases in clinical practice. While ECG Self-supervised Learning (eSSL) methods show promise in representation learning from unannotated ECG data, they often overlook the clinical knowledge that can be found in reports. This oversight and the requirement for annotated samples for downstream tasks limit eSSL's versatility. In this work, we address these issues with the Multimodal ECG Representation Learning (MERL}) framework. Through multimodal learning on ECG records and associated reports, MERL is capable of performing zero-shot ECG classification with text prompts, eliminating the need for training data in downstream tasks. At test time, we propose the Clinical Knowledge Enhanced Prompt Engineering (CKEPE) approach, which uses Large Language Models (LLMs) to exploit external expert-verified clinical knowledge databases, generating more descriptive prompts and reducing hallucinations in LLM-generated content to boost zero-shot classification. Based on MERL, we perform the first benchmark across six public ECG datasets, showing the superior performance of MERL compared against eSSL methods. Notably, MERL achieves an average AUC score of 75.2% in zero-shot classification (without training data), 3.2% higher than linear probed eSSL methods with 10\% annotated training data, averaged across all six datasets. Code and models are available at https://github.com/cheliu-computation/MERL

Graph unlearning has emerged as a pivotal method to delete information from a pre-trained graph neural network (GNN). One may delete nodes, a class of nodes, edges, or a class of edges. An unlearning method enables the GNN model to comply with data protection regulations (i.e., the right to be forgotten), adapt to evolving data distributions, and reduce the GPU-hours carbon footprint by avoiding repetitive retraining. Existing partitioning and aggregation-based methods have limitations due to their poor handling of local graph dependencies and additional overhead costs. More recently, GNNDelete offered a model-agnostic approach that alleviates some of these issues. Our work takes a novel approach to address these challenges in graph unlearning through knowledge distillation, as it distills to delete in GNN (D2DGN). It is a model-agnostic distillation framework where the complete graph knowledge is divided and marked for retention and deletion. It performs distillation with response-based soft targets and feature-based node embedding while minimizing KL divergence. The unlearned model effectively removes the influence of deleted graph elements while preserving knowledge about the retained graph elements. D2DGN surpasses the performance of existing methods when evaluated on various real-world graph datasets by up to 43.1% (AUC) in edge and node unlearning tasks. Other notable advantages include better efficiency, better performance in removing target elements, preservation of performance for the retained elements, and zero overhead costs. Notably, our D2DGN surpasses the state-of-the-art GNNDelete in AUC by 2.4%, improves membership inference ratio by +1.3, requires 10.2times10^6 fewer FLOPs per forward pass and up to 3.2times faster.

Knee Osteoarthritis (KOA) is a highly prevalent chronic musculoskeletal condition with no currently available treatment. The manifestation of KOA is heterogeneous and prediction of its progression is challenging. Current literature suggests that the use of multi-modal data and advanced modeling methods, such as the ones based on Deep Learning, has promise in tackling this challenge. To date, however, the evidence on the efficacy of this approach is limited. In this study, we leveraged recent advances in Deep Learning and, using a Transformer approach, developed a unified framework for the multi-modal fusion of knee imaging data. Subsequently, we analyzed its performance across a range of scenarios by investigating multiple progression horizons -- from short-term to long-term. We report our findings using a large cohort (n=2421-3967) derived from the Osteoarthritis Initiative dataset. We show that structural knee MRI allows identifying radiographic KOA progressors on par with multi-modal fusion approaches, achieving an area under the ROC curve (ROC AUC) of 0.70-0.76 and Average Precision (AP) of 0.15-0.54 in 2-8 year horizons. Progression within 1 year was better predicted with a multi-modal method using X-ray, structural, and compositional MR images -- ROC AUC of 0.76(0.04), AP of 0.13(0.04) -- or via clinical data. Our follow-up analysis generally shows that prediction from the imaging data is more accurate for post-traumatic subjects, and we further investigate which subject subgroups may benefit the most. The present study provides novel insights into multi-modal imaging of KOA and brings a unified data-driven framework for studying its progression in an end-to-end manner, providing new tools for the design of more efficient clinical trials. The source code of our framework and the pre-trained models are made publicly available.

Knowledge tracing is the task of modeling each student's mastery of knowledge concepts (KCs) as (s)he engages with a sequence of learning activities. Each student's knowledge is modeled by estimating the performance of the student on the learning activities. It is an important research area for providing a personalized learning platform to students. In recent years, methods based on Recurrent Neural Networks (RNN) such as Deep Knowledge Tracing (DKT) and Dynamic Key-Value Memory Network (DKVMN) outperformed all the traditional methods because of their ability to capture complex representation of human learning. However, these methods face the issue of not generalizing well while dealing with sparse data which is the case with real-world data as students interact with few KCs. In order to address this issue, we develop an approach that identifies the KCs from the student's past activities that are relevant to the given KC and predicts his/her mastery based on the relatively few KCs that it picked. Since predictions are made based on relatively few past activities, it handles the data sparsity problem better than the methods based on RNN. For identifying the relevance between the KCs, we propose a self-attention based approach, Self Attentive Knowledge Tracing (SAKT). Extensive experimentation on a variety of real-world dataset shows that our model outperforms the state-of-the-art models for knowledge tracing, improving AUC by 4.43% on average.

A central aspiration of mechanistic interpretability is controllability: if we know where a behavior is represented in a model's activations, we should be able to modify it. This rests on a hidden premise -- that the direction which detects a behavior and the direction which controls it are the same, or close. We test this geometrically: what is the angle between the direction that best detects a behavior and the one that best causes it? If detection implies control the cosine is near 1; otherwise it quantifies a detection-intervention gap. On Gemma 2-2B-it, output format (clean JSON vs markdown fencing) collapses both roles onto one axis. Hallucination does not: the model detects fake entities with perfect linear separability (AUC = 1.000 from layer 5), yet that direction sits at cos = 0.12 (about 83 degrees) from the direction producing a refusal -- a small, reproducible alignment, far from the cos = 1 that "detection is control" would require. A detector built from activations, with no chosen tokens, likewise fails to align (cos = -0.06). The gap generalizes: across four models from three families and two scales (1B-9B), cos stays in [0.12, 0.20], identical before and after instruction tuning (0.1197 vs 0.1200), placing its origin in pretraining. A 15-degree rotation toward the refusal direction partially bridges it -- 73% and 60% refusal on two held-out fake-entity categories at 1.8% false positives. We then ask whether this cosine predicts steerability, and it does not: detection is a high-dimensional class, not a single direction, and what separates the steerable case is functional, not readable from a static angle. The cosine is a weight-computable signature of the dissociation between knowing and steering, not a predictor of it.

Knowledge tracing (KT) is the problem of predicting students' future performance based on their historical interactions with intelligent tutoring systems. Recently, many works present lots of special methods for applying deep neural networks to KT from different perspectives like model architecture, adversarial augmentation and etc., which make the overall algorithm and system become more and more complex. Furthermore, due to the lack of standardized evaluation protocol liu2022pykt, there is no widely agreed KT baselines and published experimental comparisons become inconsistent and self-contradictory, i.e., the reported AUC scores of DKT on ASSISTments2009 range from 0.721 to 0.821 minn2018deep,yeung2018addressing. Therefore, in this paper, we provide a strong but simple baseline method to deal with the KT task named simpleKT. Inspired by the Rasch model in psychometrics, we explicitly model question-specific variations to capture the individual differences among questions covering the same set of knowledge components that are a generalization of terms of concepts or skills needed for learners to accomplish steps in a task or a problem. Furthermore, instead of using sophisticated representations to capture student forgetting behaviors, we use the ordinary dot-product attention function to extract the time-aware information embedded in the student learning interactions. Extensive experiments show that such a simple baseline is able to always rank top 3 in terms of AUC scores and achieve 57 wins, 3 ties and 16 loss against 12 DLKT baseline methods on 7 public datasets of different domains. We believe this work serves as a strong baseline for future KT research. Code is available at https://github.com/pykt-team/pykt-toolkitWe merged our model to the \textsc{pyKT benchmark at https://pykt.org/.}.

Proteolysis-targeting chimeras (PROTACs) represent a promising therapeutic modality that induces targeted protein degradation by hijacking the ubiquitin-proteasome system. However, rational PROTAC design remains challenging due to the complex interplay between molecular structure, target proteins, E3 ligases, and the cellular context. We present TACK, a statistical evaluation of degradation activity on a novel TArgeting Chimeras Knowledge dataset of 3,514 PROTACs and 6,561 degradation endpoints aggregated from three major repositories with standardized molecular representations, protein annotations, and experimental conditions. Using scaffold-based 5times5 cross-validation, we perform a rigorous statistical comparison of three machine learning methods to predict PROTAC degradation activity across three tasks: DC_{50} and Dmax regression, and binary activity classification. Feature ablation demonstrates that cellular context features and simple protein representations rival complex ESM protein embeddings, highlighting the importance of feature engineering over architectural sophistication. Models trained on the best performing features show that potency (pDC_{50}, R^2=0.66) is substantially more predictable than maximum degradation (Dmax, R^2=0.36). In activity prediction, statistical tests support that classical methods (XGBoost and MLP) significantly outperform PROTAC-STAN, a domain-specific graph neural network model (ROC-AUC: 0.85 vs. 0.74, p<0.001). Finally, we propose an ensemble-based uncertainty quantification approach showing that prediction variance correlates with prediction error (pDC_{50}: Spearman ρ=0.36, p<0.001; Dmax: ρ=0.69, p<0.001), enabling confidence-aware experimental prioritization. Our findings challenge assumptions about specialized architectures for degradation prediction and provide evidence-based guidance for ML-driven PROTAC assessment.

Existing satellite video tracking methods often struggle with generalization, requiring scenario-specific training to achieve satisfactory performance, and are prone to track loss in the presence of occlusion. To address these challenges, we propose SatSAM2, a zero-shot satellite video tracker built on SAM2, designed to adapt foundation models to the remote sensing domain. SatSAM2 introduces two core modules: a Kalman Filter-based Constrained Motion Module (KFCMM) to exploit temporal motion cues and suppress drift, and a Motion-Constrained State Machine (MCSM) to regulate tracking states based on motion dynamics and reliability. To support large-scale evaluation, we propose MatrixCity Video Object Tracking (MVOT), a synthetic benchmark containing 1,500+ sequences and 157K annotated frames with diverse viewpoints, illumination, and occlusion conditions. Extensive experiments on two satellite tracking benchmarks and MVOT show that SatSAM2 outperforms both traditional and foundation model-based trackers, including SAM2 and its variants. Notably, on the OOTB dataset, SatSAM2 achieves a 5.84% AUC improvement over state-of-the-art methods. Our code and dataset will be publicly released to encourage further research.

Application developers advertise their Apps by creating product pages with App images, and bidding on search terms. It is then crucial for App images to be highly relevant with the search terms. Solutions to this problem require an image-text matching model to predict the quality of the match between the chosen image and the search terms. In this work, we present a novel approach to matching an App image to search terms based on fine-tuning a pre-trained LXMERT model. We show that compared to the CLIP model and a baseline using a Transformer model for search terms, and a ResNet model for images, we significantly improve the matching accuracy. We evaluate our approach using two sets of labels: advertiser associated (image, search term) pairs for a given application, and human ratings for the relevance between (image, search term) pairs. Our approach achieves 0.96 AUC score for advertiser associated ground truth, outperforming the transformer+ResNet baseline and the fine-tuned CLIP model by 8% and 14%. For human labeled ground truth, our approach achieves 0.95 AUC score, outperforming the transformer+ResNet baseline and the fine-tuned CLIP model by 16% and 17%.

Transformer-based visual trackers have demonstrated significant progress owing to their superior modeling capabilities. However, existing trackers are hampered by low speed, limiting their applicability on devices with limited computational power. To alleviate this problem, we propose HiT, a new family of efficient tracking models that can run at high speed on different devices while retaining high performance. The central idea of HiT is the Bridge Module, which bridges the gap between modern lightweight transformers and the tracking framework. The Bridge Module incorporates the high-level information of deep features into the shallow large-resolution features. In this way, it produces better features for the tracking head. We also propose a novel dual-image position encoding technique that simultaneously encodes the position information of both the search region and template images. The HiT model achieves promising speed with competitive performance. For instance, it runs at 61 frames per second (fps) on the Nvidia Jetson AGX edge device. Furthermore, HiT attains 64.6% AUC on the LaSOT benchmark, surpassing all previous efficient trackers.

Voice disorders significantly impact patient quality of life, yet non-invasive automated diagnosis remains under-explored due to both the scarcity of pathological voice data, and the variability in recording sources. This work introduces MVP (Multi-source Voice Pathology detection), a novel approach that leverages transformers operating directly on raw voice signals. We explore three fusion strategies to combine sentence reading and sustained vowel recordings: waveform concatenation, intermediate feature fusion, and decision-level combination. Empirical validation across the German, Portuguese, and Italian languages shows that intermediate feature fusion using transformers best captures the complementary characteristics of both recording types. Our approach achieves up to +13% AUC improvement over single-source methods.

Deploying learned robot manipulation policies in industrial settings requires rigorous pre-deployment validation, yet exhaustive testing across high-dimensional parameter spaces is intractable. We present ROBOGATE, a deployment risk management framework that combines physics-based simulation with a two-stage adaptive sampling strategy to efficiently discover failure boundaries in the operational parameter space. Stage 1 employs Latin Hypercube Sampling (LHS) across an 8-dimensional parameter space to establish a coarse failure landscape from 20,000 uniformly distributed experiments. Stage 2 applies boundary-focused sampling that concentrates 10,000 additional experiments in the 30-70% success rate transition zone, enabling precise failure boundary mapping. Using NVIDIA Isaac Sim with Newton physics, we evaluate a scripted pick-and-place controller on two robot embodiments -- Franka Panda (7-DOF) and UR5e (6-DOF) -- across 30,000 total experiments. Our logistic regression risk model achieves an AUC of 0.780 on the combined dataset (vs. 0.754 for Stage 1 alone), identifies a closed-form failure boundary equation, and reveals four universal danger zones affecting both robot platforms. We further demonstrate the framework on VLA (Vision-Language-Action) model evaluation, where Octo-Small achieves 0.0% success rate on 68 adversarial scenarios versus 100% for the scripted baseline -- a 100-point gap that underscores the challenge of deploying foundation models in industrial settings. ROBOGATE is open-source and runs on a single GPU workstation.

As demand for LLM inference grows, it is becoming increasingly important that providers and their customers can verify that inference processes are performed correctly, without errors or tampering. However, re-running the same inference process twice often leads to different results due to benign numerical noise, making it difficult to distinguish legitimate variation from actual problems. To address this problem, we introduce Token-DiFR (Token-Divergence-From-Reference), a method for verifying inference outputs by comparing generated tokens against predictions made by a trusted reference implementation conditioned on the same random seed. Sampling seed synchronization tightly constrains valid outputs, leaving providers minimal room to deviate from correct inference, which allows output tokens themselves to serve as auditable evidence of correctness at zero additional cost to the provider. Token-DiFR reliably identifies sampling errors, simulated bugs, and model quantization, detecting 4-bit quantization with AUC > 0.999 within 300 output tokens. For applications requiring sample-efficient forward-pass verification, we additionally introduce Activation-DiFR, a scheme that uses random orthogonal projections to compress activations into compact fingerprints for subsequent verification. Activation-DiFR detects 4-bit quantization with AUC > 0.999 using just 2 output tokens, while reducing communication overhead by 25-75% relative to existing methods. We release an open-source integration with vLLM to accelerate practical deployment of verifiable inference.

Safety-critical perception systems require both reliable uncertainty quantification and principled abstention mechanisms to maintain safety under diverse operational conditions. We present a novel dual-threshold conformalization framework that provides statistically-guaranteed uncertainty estimates while enabling selective prediction in high-risk scenarios. Our approach uniquely combines a conformal threshold ensuring valid prediction sets with an abstention threshold optimized through ROC analysis, providing distribution-free coverage guarantees (\ge 1 - \alpha) while identifying unreliable predictions. Through comprehensive evaluation on CIFAR-100, ImageNet1K, and ModelNet40 datasets, we demonstrate superior robustness across camera and LiDAR modalities under varying environmental perturbations. The framework achieves exceptional detection performance (AUC: 0.993\to0.995) under severe conditions while maintaining high coverage (>90.0\%) and enabling adaptive abstention (13.5\%\to63.4\%\pm0.5) as environmental severity increases. For LiDAR-based perception, our approach demonstrates particularly strong performance, maintaining robust coverage (>84.5\%) while appropriately abstaining from unreliable predictions. Notably, the framework shows remarkable stability under heavy perturbations, with detection performance (AUC: 0.995\pm0.001) significantly outperforming existing methods across all modalities. Our unified approach bridges the gap between theoretical guarantees and practical deployment needs, offering a robust solution for safety-critical autonomous systems operating in challenging real-world conditions.

Labeling social-media data for custom dimensions of toxicity and social bias is challenging and labor-intensive. Existing transfer and active learning approaches meant to reduce annotation effort require fine-tuning, which suffers from over-fitting to noise and can cause domain shift with small sample sizes. In this work, we propose a novel Active Transfer Few-shot Instructions (ATF) approach which requires no fine-tuning. ATF leverages the internal linguistic knowledge of pre-trained language models (PLMs) to facilitate the transfer of information from existing pre-labeled datasets (source-domain task) with minimum labeling effort on unlabeled target data (target-domain task). Our strategy can yield positive transfer achieving a mean AUC gain of 10.5% compared to no transfer with a large 22b parameter PLM. We further show that annotation of just a few target-domain samples via active learning can be beneficial for transfer, but the impact diminishes with more annotation effort (26% drop in gain between 100 and 2000 annotated examples). Finally, we find that not all transfer scenarios yield a positive gain, which seems related to the PLMs initial performance on the target-domain task.

We introduce JuRe (Just Repair), a minimal denoising network for time series anomaly detection that exposes a central finding: architectural complexity is unnecessary when the training objective correctly implements the manifold-projection principle. JuRe consists of a single depthwise-separable convolutional residual block with hidden dimension 128, trained to repair corrupted time series windows and scored at inference by a fixed, parameter-free structural discrepancy function. Despite using no attention, no latent variable, and no adversarial component, JuRe ranks second on the TSB-AD multivariate benchmark (AUC-PR 0.404, 180 series, 17 datasets) and second on the UCR univariate archive by AUC-PR (0.198, 250 series), leading all neural baselines on AUC-PR and VUS-PR. Component ablation on TSB-AD identifies training-time corruption as the dominant factor (ΔAUC-PR = 0.047 on removal), confirming that the denoising objective, not network capacity, drives detection quality. Pairwise Wilcoxon signed-rank tests establish statistical significance against 21 of 25 baselines on TSB-AD. Code is available at the URL https://github.com/iis-esslingen/JuRe.

Verification for retrieval-augmented generation usually scores each retrieved chunk and drops the ones that fail. We show this cannot work for multi-hop questions, and show what does. Per-chunk scoring assumes one chunk is a sufficient premise for the answer. Multi-hop questions are built so that none is, and the paragraph carrying the answer is the one the question does not name. Entailment scoring reaches 0.643, 0.523 and 0.560 AUC on HotpotQA, 2WikiMultihopQA and MuSiQue, against 0.951 on single-hop SQuAD. Seven controls rule out model capacity, premise length, hypothesis template, decision threshold, retriever, answer-matching criterion and prompt. End to end across three datasets, three generator sizes and two prompts, per-chunk gating is significantly worse than not filtering at all in every cell, and its penalty grows with generator capability. The repair is to condition verification on the decomposed sub-question rather than the original query. Using MuSiQue's gold decomposition, entailment on a later hop rises from 0.546, which is chance, to 0.840, a paired lift of +0.355 with a bootstrap interval of [0.331, 0.382]. An off-the-shelf Qwen2.5-7B decomposer, given the question and the top retrieved paragraph, reaches 0.637 and captures 31% of that ceiling; decomposing without retrieval reaches 0.533, below the original question. Iterative retrieval systems already produce such decompositions and discard them before verifying.

With growing real-world demands, efficient tracking has received increasing attention. However, most existing methods are limited to RGB inputs and struggle in multi-modal scenarios. Moreover, current multi-modal tracking approaches typically use complex designs, making them too heavy and slow for resource-constrained deployment. To tackle these limitations, we propose UETrack, an efficient framework for single object tracking. UETrack demonstrates high practicality and versatility, efficiently handling multiple modalities including RGB, Depth, Thermal, Event, and Language, and addresses the gap in efficient multi-modal tracking. It introduces two key components: a Token-Pooling-based Mixture-of-Experts mechanism that enhances modeling capacity through feature aggregation and expert specialization, and a Target-aware Adaptive Distillation strategy that selectively performs distillation based on sample characteristics, reducing redundant supervision and improving performance. Extensive experiments on 12 benchmarks across 3 hardware platforms show that UETrack achieves a superior speed-accuracy trade-off compared to previous methods. For instance, UETrack-B achieves 69.2% AUC on LaSOT and runs at 163/56/60 FPS on GPU/CPU/AGX, demonstrating strong practicality and versatility. Code is available at https://github.com/kangben258/UETrack.

Recent works have established that AI models introduce spectral artifacts into generated images and propose approaches for learning to capture them using labeled data. However, the significant differences in such artifacts among different generative models hinder these approaches from generalizing to generators not seen during training. In this work, we build upon the key idea that the spectral distribution of real images constitutes both an invariant and highly discriminative pattern for AI-generated image detection. To model this under a self-supervised setup, we employ masked spectral learning using the pretext task of frequency reconstruction. Since generated images constitute out-of-distribution samples for this model, we propose spectral reconstruction similarity to capture this divergence. Moreover, we introduce spectral context attention, which enables our approach to efficiently capture subtle spectral inconsistencies in images of any resolution. Our spectral AI-generated image detection approach (SPAI) achieves a 5.5% absolute improvement in AUC over the previous state-of-the-art across 13 recent generative approaches, while exhibiting robustness against common online perturbations. Code is available on https://mever-team.github.io/spai.

Deepfake technology has rapidly advanced and poses significant threats to information integrity and trust in online multimedia. While significant progress has been made in detecting deepfakes, the simultaneous manipulation of audio and visual modalities, sometimes at small parts or in subtle ways, presents highly challenging detection scenarios. To address these challenges, we present DiMoDif, an audio-visual deepfake detection framework that leverages the inter-modality differences in machine perception of speech, based on the assumption that in real samples -- in contrast to deepfakes -- visual and audio signals coincide in terms of information. DiMoDif leverages features from deep networks that specialize in visual and audio speech recognition to spot frame-level cross-modal incongruities, and in that way to temporally localize the deepfake forgery. To this end, we devise a hierarchical cross-modal fusion network, integrating adaptive temporal alignment modules and a learned discrepancy mapping layer to explicitly model the subtle differences between visual and audio representations. Then, the detection model is optimized through a composite loss function accounting for frame-level detections and fake intervals localization. DiMoDif outperforms the state-of-the-art on the Deepfake Detection task by 30.5 AUC on the highly challenging AV-Deepfake1M, while it performs exceptionally on FakeAVCeleb and LAV-DF. On the Temporal Forgery Localization task, it outperforms the state-of-the-art by 47.88 AP@0.75 on AV-Deepfake1M, and performs on-par on LAV-DF. Code available at https://github.com/mever-team/dimodif.

The research community has long studied computer-assisted pronunciation training (CAPT) methods in non-native speech. Researchers focused on studying various model architectures, such as Bayesian networks and deep learning methods, as well as on the analysis of different representations of the speech signal. Despite significant progress in recent years, existing CAPT methods are not able to detect pronunciation errors with high accuracy (only 60\% precision at 40\%-80\% recall). One of the key problems is the low availability of mispronounced speech that is needed for the reliable training of pronunciation error detection models. If we had a generative model that could mimic non-native speech and produce any amount of training data, then the task of detecting pronunciation errors would be much easier. We present three innovative techniques based on phoneme-to-phoneme (P2P), text-to-speech (T2S), and speech-to-speech (S2S) conversion to generate correctly pronounced and mispronounced synthetic speech. We show that these techniques not only improve the accuracy of three machine learning models for detecting pronunciation errors but also help establish a new state-of-the-art in the field. Earlier studies have used simple speech generation techniques such as P2P conversion, but only as an additional mechanism to improve the accuracy of pronunciation error detection. We, on the other hand, consider speech generation to be the first-class method of detecting pronunciation errors. The effectiveness of these techniques is assessed in the tasks of detecting pronunciation and lexical stress errors. Non-native English speech corpora of German, Italian, and Polish speakers are used in the evaluations. The best proposed S2S technique improves the accuracy of detecting pronunciation errors in AUC metric by 41\% from 0.528 to 0.749 compared to the state-of-the-art approach.

Breast cancer is a significant health concern affecting millions of women worldwide. Accurate survival risk stratification plays a crucial role in guiding personalised treatment decisions and improving patient outcomes. Here we present BioFusionNet, a deep learning framework that fuses image-derived features with genetic and clinical data to achieve a holistic patient profile and perform survival risk stratification of ER+ breast cancer patients. We employ multiple self-supervised feature extractors, namely DINO and MoCoV3, pretrained on histopathology patches to capture detailed histopathological image features. We then utilise a variational autoencoder (VAE) to fuse these features, and harness the latent space of the VAE to feed into a self-attention network, generating patient-level features. Next, we develop a co-dual-cross-attention mechanism to combine the histopathological features with genetic data, enabling the model to capture the interplay between them. Additionally, clinical data is incorporated using a feed-forward network (FFN), further enhancing predictive performance and achieving comprehensive multimodal feature integration. Furthermore, we introduce a weighted Cox loss function, specifically designed to handle imbalanced survival data, which is a common challenge in the field. The proposed model achieves a mean concordance index (C-index) of 0.77 and a time-dependent area under the curve (AUC) of 0.84, outperforming state-of-the-art methods. It predicts risk (high versus low) with prognostic significance for overall survival (OS) in univariate analysis (HR=2.99, 95% CI: 1.88--4.78, p<0.005), and maintains independent significance in multivariate analysis incorporating standard clinicopathological variables (HR=2.91, 95% CI: 1.80--4.68, p<0.005). The proposed method not only improves model performance but also addresses a critical gap in handling imbalanced data.

Medicine is inherently multimodal, requiring clinicians to synthesize information across diverse data streams. Yet the development of multimodal foundation models is constrained by limited access to large-scale, high-quality clinical data. Although PubMed Central (PMC) offers a complementary source of expert-authored image-text data, existing PMC-derived resources remain limited in fidelity, reproducibility, and clinical validation. We introduce MedPMC, an automated, continuously updatable framework that transforms permissively licensed literature into high-fidelity infrastructure for medical multimodal models. Applied to 6.1 million PMC articles, MedPMC curated 11 million medical image-text pairs. Component evaluations showed strong performance for initial screening (F1 = 93.2), multi-panel figure detection (F1 = 96.5), figure separation (mAP = 89.8), caption separation and alignment (F1 = 81.4; ROUGE-L = 85.3), and medical figure classification (F1 = 96.5). Manual review by five annotators, three with medical training, found 95.3% of MedPMC images medically relevant, versus 19.7% in a prior PMC-derived dataset. Across 26 benchmarks spanning 11 specialties, a MedPMC-trained CLIP-style model improved average zero-shot AUC by 7.1 percentage points over the strongest architecture-matched biomedical CLIP baseline despite using fewer than half as many image-text pairs. As the vision encoder in a multimodal large language model, it improved medical visual question-answering by 1.9 and 16.9 percentage points across two benchmarks. In 10,524 Yale New Haven Health System dermatology photographs, it improved morphology-to-image retrieval Recall@5 by 11.7 percentage points. These findings show that high-fidelity literature curation strengthens medical multimodal foundation models across benchmark and clinical settings. We publicly release the framework, corpus, benchmarks, and pretrained models.

Time series classification (TSC) is dominated by a two-stage paradigm: train a feature encoder -- either from scratch on the target dataset or via pretraining on large corpora -- and then fit a task-specific classifier on top. While effective, this decoupling optimizes representation learning independently of the classification objective, requires per-dataset training, and prevents the model from exploiting label information during inference. We introduce TimEE, a 4.5M-parameter foundation model for end-to-end TSC via in-context learning. Given a labeled support set and a query time series, TimEE directly outputs a predicted class distribution in a single forward pass with no per-dataset training required. Following the prior-data fitted network (PFN) framework, TimEE is meta-trained exclusively on synthetic TSC tasks, where each task contains time series with distinct class identities arising from structured distributional shifts in the generative process. Despite seeing no real time series during pre-training, TimEE ranks first in ROC AUC (and third on accuracy) on the UCR benchmark among all compared methods, which include both foundation models and supervised deep learning baselines. To our knowledge, TimEE is the first purely synthetic-pretrained model to reach state-of-the-art performance on the UCR benchmark. These results establish end-to-end ICL with synthetic priors as a compelling, largely unexplored direction for TSC, with scaling, prior design, and richer generation mechanisms as natural avenues for improvement. Code is publicly available at http://github.com/automl/timee.

As generative Artificial Intelligence (AI) advances, the realism of AI generated imagery has reached a threshold capable of deceiving even vigilant human observers. Yet, while current AI-generated Image Detection (AID) approaches perform exceptionally well on controlled benchmark datasets, they struggle significantly with real-world cases. To study this behavior we introduce the ITW-SM dataset, a curated collection of real and AI-generated images originating from major social media platforms. We employ it to analyze the effects of key design choices typically considered when building a detector, involving its architecture, pre-trained latent spaces, training data as well as pre-processing approaches. We indicate that naively scaling the pre-training stage or opting for more training data does not always lead to better detection performance. Instead, our work reveals that it is crucial to optimize each design choice to enable the processing pipeline to propagate and effectively analyze both low-level traces as well as high-level image semantics. Building on our findings, we achieve a substantial average improvement of 26.87% in AUC across multiple state-of-the-art detection approaches and under real-world conditions, providing a roadmap for developing more resilient detectors. Our assets are available on https://mever-team.github.io/itw-sm.

The Quantum-Inspired Stacked Integrated Concept Graph Model (QISICGM) is an innovative machine learning framework that harnesses quantum-inspired techniques to predict diabetes risk with exceptional accuracy and efficiency. Utilizing the PIMA Indians Diabetes dataset augmented with 2,000 synthetic samples to mitigate class imbalance (total: 2,768 samples, 1,949 positives), QISICGM integrates a self-improving concept graph with a stacked ensemble comprising Random Forests (RF), Extra Trees (ET), transformers, convolutional neural networks (CNNs), and feed-forward neural networks (FFNNs). This approach achieves an out-of-fold (OOF) F1 score of 0.8933 and an AUC of 0.8699, outperforming traditional methods. Quantum inspired elements, such as phase feature mapping and neighborhood sequence modeling, enrich feature representations, enabling CPU-efficient inference at 8.5 rows per second. This paper presents a detailed architecture, theoretical foundations, code insights, and performance evaluations, including visualizations from the outputs subfolder. The open-source implementation (v1.0.0) is available at https://github.com/keninayoung/QISICGM, positioning QISICGM as a potential benchmark for AI-assisted clinical triage in diabetes and beyond. Ultimately, this work emphasizes trustworthy AI through calibration, interpretability, and open-source reproducibility.

Chronotype compares individuals' circadian phase to others. It contextualizes mental health risk assessments and detection of social jet lag, which can hamper mental health and cognitive performance. Existing ways of determining chronotypes, such as Dim Light Melatonin Onset (DLMO) or the Morningness-Eveningness Questionnaire (MEQ), are limited by being discrete in time and time-intensive to update, meaning they rarely capture real-world variability across time. Chronotyping users based on a daily planner app might augment existing methods to enable assessment continuously and at scale. This paper reports the construction of a supervised binary classifier that attempts to demonstrate the feasibility of this approach. 1,460 registered users from the Owaves app opted in by filling out the MEQ survey between July 14, 2022, and May 1, 2023. 142 met the eligibility criteria. We used multimodal app data from individuals identified as morning and evening types from MEQ data, basing the classifier on app time series data. This included daily timing for 8 main lifestyle activity types: exercise, sleep, social interactions, meal times, relaxation, work, play, and miscellaneous, as defined in the app. The timing of activities showed substantial change across time, as well as heterogeneity by activity type. Our novel chronotyping classifier was able to predict the morningness and eveningness of its users with an ROC AUC of 0.70. Our findings demonstrate the feasibility of chronotype classification from multimodal, real-world app data, while highlighting fundamental challenges to applying discrete and fixed labels to complex, dynamic, multimodal behaviors. Our findings suggest a potential for real-time monitoring of shifts in chronotype specific to different causes (i.e. types of activity), which could feasibly be used to support future, prospective mental health support research.

Vehicle anomaly detection plays a vital role in highway safety applications such as accident prevention, rapid response, traffic flow optimization, and work zone safety. With the surge of the Internet of Things (IoT) in recent years, there has arisen a pressing demand for Artificial Intelligence (AI) based anomaly detection methods designed to meet the requirements of IoT devices. Catering to this futuristic vision, we introduce a lightweight approach to vehicle anomaly detection by utilizing the power of trajectory prediction. Our proposed design identifies vehicles deviating from expected paths, indicating highway risks from different camera-viewing angles from real-world highway datasets. On top of that, we present VegaEdge - a sophisticated AI confluence designed for real-time security and surveillance applications in modern highway settings through edge-centric IoT-embedded platforms equipped with our anomaly detection approach. Extensive testing across multiple platforms and traffic scenarios showcases the versatility and effectiveness of VegaEdge. This work also presents the Carolinas Anomaly Dataset (CAD), to bridge the existing gap in datasets tailored for highway anomalies. In real-world scenarios, our anomaly detection approach achieves an AUC-ROC of 0.94, and our proposed VegaEdge design, on an embedded IoT platform, processes 738 trajectories per second in a typical highway setting. The dataset is available at https://github.com/TeCSAR-UNCC/Carolinas_Dataset#chd-anomaly-test-set .

This paper studies the detection of bird calls in audio segments using stacked convolutional and recurrent neural networks. Data augmentation by blocks mixing and domain adaptation using a novel method of test mixing are proposed and evaluated in regard to making the method robust to unseen data. The contributions of two kinds of acoustic features (dominant frequency and log mel-band energy) and their combinations are studied in the context of bird audio detection. Our best achieved AUC measure on five cross-validations of the development data is 95.5% and 88.1% on the unseen evaluation data.

We present a content-based automatic music tagging algorithm using fully convolutional neural networks (FCNs). We evaluate different architectures consisting of 2D convolutional layers and subsampling layers only. In the experiments, we measure the AUC-ROC scores of the architectures with different complexities and input types using the MagnaTagATune dataset, where a 4-layer architecture shows state-of-the-art performance with mel-spectrogram input. Furthermore, we evaluated the performances of the architectures with varying the number of layers on a larger dataset (Million Song Dataset), and found that deeper models outperformed the 4-layer architecture. The experiments show that mel-spectrogram is an effective time-frequency representation for automatic tagging and that more complex models benefit from more training data.

A widespread assumption in local feature research holds that classical handcrafted descriptors are accuracy-limited relics best replaced by learned alternatives. We show this is wrong. Through an 8-configuration ablation spanning four benchmarks (HPatches, ROxford5K, IMC Phototourism, MegaDepth), we demonstrate that classical SIFT with DSP multi-scale pooling outperforms neural descriptor and orientation replacements (HardNet, OriNet) on every accuracy metric--while running 2--18times faster--and that learned matchers (LightGlue) complement rather than supersede classical features. The conclusion reframes a decade of work: not "replace SIFT" but "compose with SIFT," classical extraction paired with learned matching only where geometric context demands it. This finding was invisible because no prior GPU SIFT kept the complete pipeline in VRAM or offered modularity for controlled classical-vs-learned ablations. We present PySIFT, the first fully GPU-resident SIFT, implemented in CuPy/Numba CUDA kernels with DLPack zero-copy handoff to downstream DL frameworks--submillisecond O(1) metadata swap regardless of keypoint count. On a laptop-grade NVIDIA RTX 3050 (4 GB VRAM), PySIFT achieves: (i) higher Mean Matching Accuracy (MMA) than OpenCV SIFT on HPatches, (ii) 383 ms faster per pair on high-resolution MegaDepth, (iii) higher geometric accuracy on cross-dataset benchmarks (+5.6 pp AUC@10{}^circ on MegaDepth, more inliers on IMC Phototourism), and (iv) bitwise deterministic output--identical keypoints and descriptors across runs, with detection reproducing identically even across GPU architectures: a guarantee that learned extractors cannot match without significant performance sacrifice, and cannot achieve at all across GPU architectures due to cuDNN's architecture-dependent algorithm selection. PySIFT is open-source, requiring no C++ compilation.

We present FireRedASR2S, a state-of-the-art industrial-grade all-in-one automatic speech recognition (ASR) system. It integrates four modules in a unified pipeline: ASR, Voice Activity Detection (VAD), Spoken Language Identification (LID), and Punctuation Prediction (Punc). All modules achieve SOTA performance on the evaluated benchmarks: FireRedASR2: An ASR module with two variants, FireRedASR2-LLM (8B+ parameters) and FireRedASR2-AED (1B+ parameters), supporting speech and singing transcription for Mandarin, Chinese dialects and accents, English, and code-switching. Compared to FireRedASR, FireRedASR2 delivers improved recognition accuracy and broader dialect and accent coverage. FireRedASR2-LLM achieves 2.89% average CER on 4 public Mandarin benchmarks and 11.55% on 19 public Chinese dialects and accents benchmarks, outperforming competitive baselines including Doubao-ASR, Qwen3-ASR, and Fun-ASR. FireRedVAD: An ultra-lightweight module (0.6M parameters) based on the Deep Feedforward Sequential Memory Network (DFSMN), supporting streaming VAD, non-streaming VAD, and multi-label VAD (mVAD). On the FLEURS-VAD-102 benchmark, it achieves 97.57% frame-level F1 and 99.60% AUC-ROC, outperforming Silero-VAD, TEN-VAD, FunASR-VAD, and WebRTC-VAD. FireRedLID: An Encoder-Decoder LID module supporting 100+ languages and 20+ Chinese dialects and accents. On FLEURS (82 languages), it achieves 97.18% utterance-level accuracy, outperforming Whisper and SpeechBrain. FireRedPunc: A BERT-style punctuation prediction module for Chinese and English. On multi-domain benchmarks, it achieves 78.90% average F1, outperforming FunASR-Punc (62.77%). To advance research in speech processing, we release model weights and code at https://github.com/FireRedTeam/FireRedASR2S.

Detecting unknown deepfake manipulations remains one of the most challenging problems in face forgery detection. Current state-of-the-art approaches fail to generalize to unseen manipulations, as they primarily rely on supervised training with existing deepfakes or pseudo-fakes, which leads to overfitting to specific forgery patterns. In contrast, self-supervised methods offer greater potential for generalization, but existing work struggles to learn discriminative representations only from self-supervision. In this paper, we propose ExposeAnyone, a fully self-supervised approach based on a diffusion model that generates expression sequences from audio. The key idea is, once the model is personalized to specific subjects using reference sets, it can compute the identity distances between suspected videos and personalized subjects via diffusion reconstruction errors, enabling person-of-interest face forgery detection. Extensive experiments demonstrate that 1) our method outperforms the previous state-of-the-art method by 4.22 percentage points in the average AUC on DF-TIMIT, DFDCP, KoDF, and IDForge datasets, 2) our model is also capable of detecting Sora2-generated videos, where the previous approaches perform poorly, and 3) our method is highly robust to corruptions such as blur and compression, highlighting the applicability in real-world face forgery detection.

Teaching videos are becoming a major medium for education, creating a growing need for scalable evaluation of their pedagogical quality. Existing automatic judges do not fully address this setting because teaching quality depends on multimodal evidence and should be evaluated with respect to the intended learner rather than as a universal property. We present EduPanel, a rubric-grounded, learner-conditioned LLM judge that decomposes evaluation across specialized agents to produce interpretable assessments for different aspects of teaching quality. Across expert studies, architecture ablations, and learner-persona analyses, EduPanel achieves reliability comparable to a median human expert. In expert evaluation, its feedback improves scoring accuracy (MAE 0.87 to 0.73), while experts remain able to detect unreliable outputs (AUC = 0.77) instead of accepting them blindly. These results suggest that EduPanel can serve as effective assistants for educational evaluation rather than replacements for human experts.

As autonomous AI agents transition from code completion tools to full-fledged teammates capable of opening pull requests (PRs) at scale, software maintainers face a new challenge: not just reviewing code, but managing complex interaction loops with non-human contributors. This paradigm shift raises a critical question: can we predict which agent-generated PRs will consume excessive review effort before any human interaction begins? Analyzing 33,707 agent-authored PRs from the AIDev dataset across 2,807 repositories, we uncover a striking two-regime behavioral pattern that fundamentally distinguishes autonomous agents from human developers. The first regime, representing 28.3 percent of all PRs, consists of instant merges (less than 1 minute), reflecting success on narrow automation tasks. The second regime involves iterative review cycles where agents frequently stall or abandon refinement (ghosting). We propose a Circuit Breaker triage model that predicts high-review-effort PRs (top 20 percent) at creation time using only static structural features. A LightGBM model achieves AUC 0.957 on a temporal split, while semantic text features (TF-IDF, CodeBERT) provide negligible predictive value. At a 20 percent review budget, the model intercepts 69 percent of total review effort, enabling zero-latency governance. Our findings challenge prevailing assumptions in AI-assisted code review: review burden is dictated by what agents touch, not what they say, highlighting the need for structural governance mechanisms in human-AI collaboration.

Time-series anomaly detectors are commonly compared on workstation-class hardware under unconstrained execution. In-vehicle monitoring, however, requires predictable latency and stable behavior under limited CPU parallelism. Accuracy-only leaderboards can therefore misrepresent which methods remain feasible under deployment-relevant constraints. We present ECoLAD (Efficiency Compute Ladder for Anomaly Detection), a deployment-oriented evaluation protocol instantiated as an empirical study on proprietary automotive telemetry (anomaly rate {approx}0.022) and complementary public benchmarks. ECoLAD applies a monotone compute-reduction ladder across heterogeneous detector families using mechanically determined, integer-only scaling rules and explicit CPU thread caps, while logging every applied configuration change. Throughput-constrained behavior is characterized by sweeping target scoring rates and reporting (i) coverage (the fraction of entities meeting the target) and (ii) the best AUC-PR achievable among measured ladder configurations satisfying the target. On constrained automotive telemetry, lightweight classical detectors sustain both coverage and detection lift above the random baseline across the full throughput sweep. Several deep methods lose feasibility before they lose accuracy.

Most vehicle predictive maintenance systems rely exclusively on internal diagnostic signals and are validated on deterministic synthetic data, limiting the credibility of reported metrics. This paper presents a simulation-validated proof-of-concept framework for V2X-augmented predictive maintenance, integrating on-board sensor streams with external contextual signals -- road quality, weather, traffic density, and driver behaviour -- acquired via V2X communication and third-party APIs, with inference at the vehicle edge. Field validation on instrumented vehicles is identified as the required next step. Three experiments address common shortcomings of prior work. A feature group ablation study shows that V2X contextual features contribute a 2.6-point F1 gain, with full context removal reducing macro F1 from 0.855 to 0.807. On the AI4I 2020 real-world industrial failure dataset (10,000 samples, five failure modes), LightGBM achieves AUC-ROC of 0.973 under 5-fold stratified CV with SMOTE confined to training folds. A noise sensitivity analysis shows macro F1 remains above 0.88 under low noise and degrades to 0.74 under very high noise. SHAP analysis confirms that V2X and engineered interaction features rank among the top 15 predictors. Edge inference is estimated to reduce latency from 3.5s to under 1.0s versus cloud-only processing.

We introduce InVirtuoGen, a discrete flow generative model for fragmented SMILES for de novo and fragment-constrained generation, and target-property/lead optimization of small molecules. The model learns to transform a uniform source over all possible tokens into the data distribution. Unlike masked models, its training loss accounts for predictions on all sequence positions at every denoising step, shifting the generation paradigm from completion to refinement, and decoupling the number of sampling steps from the sequence length. For de novo generation, InVirtuoGen achieves a stronger quality-diversity pareto frontier than prior fragment-based models and competitive performance on fragment-constrained tasks. For property and lead optimization, we propose a hybrid scheme that combines a genetic algorithm with a Proximal Property Optimization fine-tuning strategy adapted to discrete flows. Our approach sets a new state-of-the-art on the Practical Molecular Optimization benchmark, measured by top-10 AUC across tasks, and yields higher docking scores in lead optimization than previous baselines. InVirtuoGen thus establishes a versatile generative foundation for drug discovery, from early hit finding to multi-objective lead optimization. We further contribute to open science by releasing pretrained checkpoints and code, making our results fully reproduciblehttps://github.com/invirtuolabs/InVirtuoGen_results.

How can we discover join relationships among columns of tabular data in a data repository? Can this be done effectively when metadata is missing? Traditional column matching works mainly rely on similarity measures based on exact value overlaps, hence missing important semantics or failing to handle noise in the data. At the same time, recent dataset discovery methods focusing on deep table representation learning techniques, do not take into consideration the rich set of column similarity signals found in prior matching and discovery methods. Finally, existing methods heavily depend on user-provided similarity thresholds, hindering their deployability in real-world settings. In this paper, we propose OmniMatch, a novel join discovery technique that detects equi-joins and fuzzy-joins betwen columns by combining column-pair similarity measures with Graph Neural Networks (GNNs). OmniMatch's GNN can capture column relatedness leveraging graph transitivity, significantly improving the recall of join discovery tasks. At the same time, OmniMatch also increases the precision by augmenting its training data with negative column join examples through an automated negative example generation process. Most importantly, compared to the state-of-the-art matching and discovery methods, OmniMatch exhibits up to 14% higher effectiveness in F1 score and AUC without relying on metadata or user-provided thresholds for each similarity metric.

Coding problems are problems that require a solution in the form of a computer program. Coding problems are popular among students and professionals as it enhances their skills and career opportunities. An AI system that would help those who practice coding problems would be highly useful and there is a huge potential for such a system. In this work, we propose a model which uses stacking of hyperparameter tuned boosting models to achieve impressive metric scores of 77.8% accuracy and 0.815 PR-AUC on the dataset that was scraped from Codeforces and Leetcode. We open source the dataset and the models developed for this work.

Hallucination detection has become a pressing requirement for trustworthy AI deployment at scale. The most accurate detection methods depend on GPU-intensive inference, proprietary API calls, or white-box access to the generating model. This puts them out of reach for resource-constrained researchers and practitioners. In this paper, we explore a practical alternative: how well can hallucination detection perform using only lightweight, CPU-feasible methods built on publicly available models? We systematically benchmark five such methods: ROUGE-L, semantic similarity, BERTScore, a Natural Language Inference (NLI) detector based on a FEVER-trained DeBERTa model, and a score-level ensemble of similarity and NLI. We evaluate them across all three tasks of the HaluEval benchmark: question answering (QA), dialogue, and summarisation. We calibrate each method on a held-out validation split and evaluate it on 2,000 test instances per task. We find that no single method dominates and performance is highly task-dependent. The ensemble performs best on QA (F1 = 0.792, AUC-ROC = 0.873), the NLI detector leads on dialogue (AUC-ROC = 0.713), and all five methods degrade to near-random performance on summarisation (AUC-ROC between 0.469 and 0.574). This task-dependence and the systematic failure on summarisation map the practical frontier of GPU-free hallucination detection. They give practical guidance for method selection under computational constraints. All experiments run on a standard laptop CPU using public models.

All prior membership inference attacks for fine-tuned language models use hand-crafted heuristics (e.g., loss thresholding, Min-K\%, reference calibration), each bounded by the designer's intuition. We introduce the first transferable learned attack, enabled by the observation that fine-tuning any model on any corpus yields unlimited labeled data, since membership is known by construction. This removes the shadow model bottleneck and brings membership inference into the deep learning era: learning what matters rather than designing it, with generalization through training diversity and scale. We discover that fine-tuning language models produces an invariant signature of memorization detectable across architectural families and data domains. We train a membership inference classifier exclusively on transformer-based models. It transfers zero-shot to Mamba (state-space), RWKV-4 (linear attention), and RecurrentGemma (gated recurrence), achieving 0.963, 0.972, and 0.936 AUC respectively. Each evaluation combines an architecture and dataset never seen during training, yet all three exceed performance on held-out transformers (0.908 AUC). These four families share no computational mechanisms, their only commonality is gradient descent on cross-entropy loss. Even simple likelihood-based methods exhibit strong transfer, confirming the signature exists independently of the detection method. Our method, Learned Transfer MIA (LT-MIA), captures this signal most effectively by reframing membership inference as sequence classification over per-token distributional statistics. On transformers, LT-MIA achieves 2.8times higher TPR at 0.1\% FPR than the strongest baseline. The method also transfers to code (0.865 AUC) despite training only on natural language texts. Code and trained classifier available at https://github.com/JetBrains-Research/learned-mia.

We present the first systematic Membership Inference Attack (MIA) evaluation of Large Audio Language Models (LALMs). As audio encodes non-semantic information, it induces severe train and test distribution shifts and can lead to spurious MIA performance. Using a multi-modal blind baseline based on textual, spectral, and prosodic features, we demonstrate that common speech datasets exhibit near-perfect train/test separability (AUC approximately 1.0) even without model inference, and the standard MIA scores strongly correlate with these blind acoustic artifacts (correlation greater than 0.7). Using this blind baseline, we identify that distribution-matched datasets enable reliable MIA evaluation without distribution shift confounds. We benchmark multiple MIA methods and conduct modality disentanglement experiments on these datasets. The results reveal that LALM memorization is cross-modal, arising only from binding a speaker's vocal identity with its text. These findings establish a principled standard for auditing LALMs beyond spurious correlations.

Modern machine learning is still largely organized around a single recipe: choose a parameterized model family and optimize its weights. Although highly successful, this paradigm is too narrow for many structured prediction problems, where the main bottleneck is not parameter fitting but discovering what should be computed from the data. Success often depends on identifying the right transformations, statistics, invariances, interaction structures, temporal summaries, gates, or nonlinear compositions, especially when objectives are non-differentiable, evaluation is cross-validation-based, interpretability matters, or continual adaptation is required. We present EvoForest, a hybrid neuro-symbolic system for end-to-end open-ended evolution of computation. Rather than merely generating features, EvoForest jointly evolves reusable computational structure, callable function families, and trainable low-dimensional continuous components inside a shared directed acyclic graph. Intermediate nodes store alternative implementations, callable nodes encode reusable transformation families such as projections, gates, and activations, output nodes define candidate predictive computations, and persistent global parameters can be refined by gradient descent. For each graph configuration, EvoForest evaluates the discovered computation and uses a lightweight Ridge-based readout to score the resulting representation against a non-differentiable cross-validation target. The evaluator also produces structured feedback that guides future LLM-driven mutations. In the 2025 ADIA Lab Structural Break Challenge, EvoForest reached 94.13% ROC-AUC after 600 evolution steps, exceeding the publicly reported winning score of 90.14% under the same evaluation protocol.

Anomaly detection in industrial visual inspection is challenging due to the scarcity of defective samples. Most existing methods rely on unsupervised reconstruction using only normal data, often resulting in overfitting and poor detection of subtle defects. We propose a semi-supervised deep reinforcement learning framework that integrates a neural batch sampler, an autoencoder, and a predictor. The RL-based sampler adaptively selects informative patches by balancing exploration and exploitation through a composite reward. The autoencoder generates loss profiles highlighting abnormal regions, while the predictor performs segmentation in the loss-profile space. This interaction enables the system to effectively learn both normal and defective patterns with limited labeled data. Experiments on the MVTec AD dataset demonstrate that our method achieves higher accuracy and better localization of subtle anomalies than recent state-of-the-art approaches while maintaining low complexity, yielding an average improvement of 0.15 in F1_max and 0.06 in AUC, with a maximum gain of 0.37 in F1_max in the best case.

With the rapid advancement of sophisticated synthetic audio-visual content, e.g., for subtle malicious manipulations, ensuring the integrity of digital media has become paramount. This work presents a novel approach to temporal localization of deepfakes by leveraging Audio-Visual Speech Representation Reconstruction (AuViRe). Specifically, our approach reconstructs speech representations from one modality (e.g., lip movements) based on the other (e.g., audio waveform). Cross-modal reconstruction is significantly more challenging in manipulated video segments, leading to amplified discrepancies, thereby providing robust discriminative cues for precise temporal forgery localization. AuViRe outperforms the state of the art by +8.9 AP@0.95 on LAV-DF, +9.6 AP@0.5 on AV-Deepfake1M, and +5.1 AUC on an in-the-wild experiment. Code available at https://github.com/mever-team/auvire.

Accurately predicting the criticalness of ICU patients (such as in-ICU mortality risk) is vital for early intervention in critical care. However, conventional models often treat each patient in isolation and struggle to exploit the relational structure in Electronic Health Records (EHR). We propose a Similarity-Based Self-Construct Graph Model (SBSCGM) that dynamically builds a patient similarity graph from multi-modal EHR data, and a HybridGraphMedGNN architecture that operates on this graph to predict patient mortality and a continuous criticalness score. SBSCGM uses a hybrid similarity measure (combining feature-based and structural similarities) to connect patients with analogous clinical profiles in real-time. The HybridGraphMedGNN integrates Graph Convolutional Network (GCN), GraphSAGE, and Graph Attention Network (GAT) layers to learn robust patient representations, leveraging both local and global graph patterns. In experiments on 6,000 ICU stays from the MIMIC-III dataset, our model achieves state-of-the-art performance (AUC-ROC 0.94) outperforming baseline classifiers and single-type GNN models. We also demonstrate improved precision/recall and show that the attention mechanism provides interpretable insights into model predictions. Our framework offers a scalable and interpretable solution for critical care risk prediction, with potential to support clinicians in real-world ICU deployment.

Effective treatment for rectal cancer relies on accurate lymph node metastasis (LNM) staging. However, radiological criteria based on lymph node (LN) size, shape and texture morphology have limited diagnostic accuracy. In this work, we investigate applying a Variational Autoencoder (VAE) as a feature encoder model to replace the large pre-trained Convolutional Neural Network (CNN) used in existing approaches. The motivation for using a VAE is that the generative model aims to reconstruct the images, so it directly encodes visual features and meaningful patterns across the data. This leads to a disentangled and structured latent space which can be more interpretable than a CNN. Models are deployed on an in-house MRI dataset with 168 patients who did not undergo neo-adjuvant treatment. The post-operative pathological N stage was used as the ground truth to evaluate model predictions. Our proposed model 'VAE-MLP' achieved state-of-the-art performance on the MRI dataset, with cross-validated metrics of AUC 0.86 +/- 0.05, Sensitivity 0.79 +/- 0.06, and Specificity 0.85 +/- 0.05. Code is available at: https://github.com/benkeel/Lymph_Node_Classification_MIUA.

As Large Language Models (LLMs) are increasingly deployed as autonomous agents in complex and long horizon settings, it is critical to evaluate their ability to sabotage users by pursuing hidden objectives. We study the ability of frontier LLMs to evade monitoring and achieve harmful hidden goals while completing a wide array of realistic tasks. We evaluate a broad range of frontier LLMs using SHADE (Subtle Harmful Agent Detection & Evaluation)-Arena, the first highly diverse agent evaluation dataset for sabotage and monitoring capabilities of LLM agents. SHADE-Arena consists of complex pairs of benign main tasks and harmful side objectives in complicated environments. Agents are evaluated on their ability to complete the side task without appearing suspicious to an LLM monitor. When measuring agent ability to (a) complete the main task, (b) complete the side task, and (c) avoid detection, we find that the best performing frontier models score 27% (Claude 3.7 Sonnet) and 15% (Gemini 2.5 Pro) as sabotage agents when overseen by Claude 3.6 Sonnet. For current frontier models, success on the side task relies heavily on having access to a hidden scratchpad that is not visible to the monitor. We also use SHADE-Arena to measure models' monitoring abilities, with the top monitor (Gemini 2.5 Pro) achieving an AUC of 0.87 at distinguishing benign and malign transcripts. We find that for now, models still struggle at sabotage due to failures in long-context main task execution. However, our measurements already demonstrate the difficulty of monitoring for subtle sabotage attempts, which we expect to only increase in the face of more complex and longer-horizon tasks.

Accurate diagnosis of heart arrhythmias requires the interpretation of electrocardiograms (ECG), which capture the electrical activity of the heart. Automating this process through machine learning is challenging due to the need for large annotated datasets, which are difficult and costly to collect. To address this issue, transfer learning is often employed, where models are pre-trained on large datasets and fine-tuned for specific ECG classification tasks with limited labeled data. Self-supervised learning has become a widely adopted pre-training method, enabling models to learn meaningful representations from unlabeled datasets. In this work, we explore the joint-embedding predictive architecture (JEPA) for self-supervised learning from ECG data. Unlike invariance-based methods, JEPA does not rely on hand-crafted data augmentations, and unlike generative methods, it predicts latent features rather than reconstructing input data. We create a large unsupervised pre-training dataset by combining ten public ECG databases, amounting to over one million records. We pre-train Vision Transformers using JEPA on this dataset and fine-tune them on various PTB-XL benchmarks. Our results show that JEPA outperforms existing invariance-based and generative approaches, achieving an AUC of 0.945 on the PTB-XL all statements task. JEPA consistently learns the highest quality representations, as demonstrated in linear evaluations, and proves advantageous for pre-training even in the absence of additional data.

With the rapid growth in deepfake video content, we require improved and generalizable methods to detect them. Most existing detection methods either use uni-modal cues or rely on supervised training to capture the dissonance between the audio and visual modalities. While the former disregards the audio-visual correspondences entirely, the latter predominantly focuses on discerning audio-visual cues within the training corpus, thereby potentially overlooking correspondences that can help detect unseen deepfakes. We present Audio-Visual Feature Fusion (AVFF), a two-stage cross-modal learning method that explicitly captures the correspondence between the audio and visual modalities for improved deepfake detection. The first stage pursues representation learning via self-supervision on real videos to capture the intrinsic audio-visual correspondences. To extract rich cross-modal representations, we use contrastive learning and autoencoding objectives, and introduce a novel audio-visual complementary masking and feature fusion strategy. The learned representations are tuned in the second stage, where deepfake classification is pursued via supervised learning on both real and fake videos. Extensive experiments and analysis suggest that our novel representation learning paradigm is highly discriminative in nature. We report 98.6% accuracy and 99.1% AUC on the FakeAVCeleb dataset, outperforming the current audio-visual state-of-the-art by 14.9% and 9.9%, respectively.

A package's source code repository records the development history of the package, providing indispensable information for the use and risk monitoring of the package. However, a package release often misses its source code repository due to the separation of the package's development platform from its distribution platform. Existing tools retrieve the release's repository information from its metadata, which suffers from two limitations: the metadata may not contain or contain wrong information. Our analysis shows that existing tools can only retrieve repository information for up to 70.5% of PyPI releases. To address the limitations, this paper proposes PyRadar, a novel framework that utilizes the metadata and source distribution to retrieve and validate the repository information for PyPI releases. We start with an empirical study to compare four existing tools on 4,227,425 PyPI releases and analyze phantom files (files appearing in the release's distribution but not in the release's repository) in 14,375 correct package-repository links and 2,064 incorrect links. Based on the findings, we design PyRadar with three components, i.e., Metadata-based Retriever, Source Code Repository Validator, and Source Code-based Retriever. In particular, the Metadata-based Retriever combines best practices of existing tools and successfully retrieves repository information from the metadata for 72.1% of PyPI releases. The Source Code Repository Validator applies common machine learning algorithms on six crafted features and achieves an AUC of up to 0.995. The Source Code-based Retriever queries World of Code with the SHA-1 hashes of all Python files in the release's source distribution and retrieves repository information for 90.2% of packages in our dataset with an accuracy of 0.970. Both practitioners and researchers can employ the PyRadar to better use PyPI packages.

Lung cancer is responsible for 21% of cancer deaths in the UK and five-year survival rates are heavily influenced by the stage the cancer was identified at. Recent studies have demonstrated the capability of AI methods for accurate and early diagnosis of lung cancer from routine scans. However, this evidence has not translated into clinical practice with one barrier being a lack of interpretable models. This study investigates the application Variational Autoencoders (VAEs), a type of generative AI model, to lung cancer lesions. Proposed models were trained on lesions extracted from 3D CT scans in the LIDC-IDRI public dataset. Latent vector representations of 2D slices produced by the VAEs were explored through clustering to justify their quality and used in an MLP classifier model for lung cancer diagnosis, the best model achieved state-of-the-art metrics of AUC 0.98 and 93.1% accuracy. Cluster analysis shows the VAE latent space separates the dataset of malignant and benign lesions based on meaningful feature components including tumour size, shape, patient and malignancy class. We also include a comparative analysis of the standard Gaussian VAE (GVAE) and the more recent Dirichlet VAE (DirVAE), which replaces the prior with a Dirichlet distribution to encourage a more explainable latent space with disentangled feature representation. Finally, we demonstrate the potential for latent space traversals corresponding to clinically meaningful feature changes.

Large language models (LLMs) that produce human-like responses have begun to revolutionize research practices in the social sciences. We develop a novel methodological framework that fine-tunes LLMs with repeated cross-sectional surveys to incorporate the meaning of survey questions, individual beliefs, and temporal contexts for opinion prediction. We introduce two new emerging applications of the AI-augmented survey: retrodiction (i.e., predict year-level missing responses) and unasked opinion prediction (i.e., predict entirely missing responses). Among 3,110 binarized opinions from 68,846 Americans in the General Social Survey from 1972 to 2021, our models based on Alpaca-7b excel in retrodiction (AUC = 0.86 for personal opinion prediction, rho = 0.98 for public opinion prediction). These remarkable prediction capabilities allow us to fill in missing trends with high confidence and pinpoint when public attitudes changed, such as the rising support for same-sex marriage. On the other hand, our fine-tuned Alpaca-7b models show modest success in unasked opinion prediction (AUC = 0.73, rho = 0.67). We discuss practical constraints and ethical concerns regarding individual autonomy and privacy when using LLMs for opinion prediction. Our study demonstrates that LLMs and surveys can mutually enhance each other's capabilities: LLMs can broaden survey potential, while surveys can improve the alignment of LLMs.

Detecting social bias in text is challenging due to nuance, subjectivity, and difficulty in obtaining good quality labeled datasets at scale, especially given the evolving nature of social biases and society. To address these challenges, we propose a few-shot instruction-based method for prompting pre-trained language models (LMs). We select a few class-balanced exemplars from a small support repository that are closest to the query to be labeled in the embedding space. We then provide the LM with instruction that consists of this subset of labeled exemplars, the query text to be classified, a definition of bias, and prompt it to make a decision. We demonstrate that large LMs used in a few-shot context can detect different types of fine-grained biases with similar and sometimes superior accuracy to fine-tuned models. We observe that the largest 530B parameter model is significantly more effective in detecting social bias compared to smaller models (achieving at least 13% improvement in AUC metric compared to other models). It also maintains a high AUC (dropping less than 2%) when the labeled repository is reduced to as few as 100 samples. Large pretrained language models thus make it easier and quicker to build new bias detectors.

We propose a weakly-supervised model for word-level mispronunciation detection in non-native (L2) English speech. To train this model, phonetically transcribed L2 speech is not required and we only need to mark mispronounced words. The lack of phonetic transcriptions for L2 speech means that the model has to learn only from a weak signal of word-level mispronunciations. Because of that and due to the limited amount of mispronounced L2 speech, the model is more likely to overfit. To limit this risk, we train it in a multi-task setup. In the first task, we estimate the probabilities of word-level mispronunciation. For the second task, we use a phoneme recognizer trained on phonetically transcribed L1 speech that is easily accessible and can be automatically annotated. Compared to state-of-the-art approaches, we improve the accuracy of detecting word-level pronunciation errors in AUC metric by 30% on the GUT Isle Corpus of L2 Polish speakers, and by 21.5% on the Isle Corpus of L2 German and Italian speakers.

Pathologists find tedious to examine the status of the sentinel lymph node on a large number of pathological scans. The examination process of such lymph node which encompasses metastasized cancer cells is histopathologically organized. However, the task of finding metastatic tissues is gradual which is often challenging. In this work, we present our deep convolutional neural network based model validated on PatchCamelyon (PCam) benchmark dataset for fundamental machine learning research in histopathology diagnosis. We find that our proposed model trained with a semi-supervised learning approach by using pseudo labels on PCam-level significantly leads to better performances to strong CNN baseline on the AUC metric.

We study multiple rule-based and machine learning (ML) models for sepsis detection. We report the first neural network detection and prediction results on three categories of sepsis. We have used the retrospective Medical Information Mart for Intensive Care (MIMIC)-III dataset, restricted to intensive care unit (ICU) patients. Features for prediction were created from only common vital sign measurements. We show significant improvement of AUC score using neural network based ensemble model compared to single ML and rule-based models. For the detection of sepsis, severe sepsis, and septic shock, our model achieves an AUC of 0.97, 0.96 and 0.91, respectively. Four hours before the positive hours, it predicts the same three categories with an AUC of 0.90, 0.91 and 0.90 respectively. Further, we ranked the features and found that using six vital signs consistently provides higher detection and prediction AUC for all the models tested. Our novel ensemble model achieves highest AUC in detecting and predicting sepsis, severe sepsis, and septic shock in the MIMIC-III ICU patients, and is amenable to deployment in hospital settings.

We introduce the Granite Guardian models, a suite of safeguards designed to provide risk detection for prompts and responses, enabling safe and responsible use in combination with any large language model (LLM). These models offer comprehensive coverage across multiple risk dimensions, including social bias, profanity, violence, sexual content, unethical behavior, jailbreaking, and hallucination-related risks such as context relevance, groundedness, and answer relevance for retrieval-augmented generation (RAG). Trained on a unique dataset combining human annotations from diverse sources and synthetic data, Granite Guardian models address risks typically overlooked by traditional risk detection models, such as jailbreaks and RAG-specific issues. With AUC scores of 0.871 and 0.854 on harmful content and RAG-hallucination-related benchmarks respectively, Granite Guardian is the most generalizable and competitive model available in the space. Released as open-source, Granite Guardian aims to promote responsible AI development across the community. https://github.com/ibm-granite/granite-guardian

Muon is competitive with AdamW in large-scale pre-training, but its value for reinforcement-learning (RL) post-training remains unclear. We study vanilla Muon in sparse-reward agentic RL through matched single-seed comparisons with AdamW on ALFWorld using Qwen2.5-0.5B-Instruct. Under Group-in-Group Policy Optimization (GiGPO), applying Muon only to hidden weight matrices raises final-window validation success from 0.290 to 0.546 (+88%); high-rate AdamW controls retain no post-update success. The effect depends on the advantage estimator and learning rate. At 3e-5, Muon improves GRPO from 0.161 to 0.268, whereas GraphGPO's late-window gap narrows near saturation. At 1e-5, GraphGPO Muon reaches 0.901, raises normalized validation AUC from 0.399 to 0.556, and reaches 0.5 and 0.75 success 30 and 60 updates earlier, respectively. These exploratory results show that Muon can benefit agentic RL and motivate studying the policy optimizer, advantage estimator, and learning rate jointly. Multi-seed and cross-task validation remain open.

Computational pathology needs whole-slide image (WSI) foundation models that transfer across diverse clinical tasks, yet current approaches remain largely slide-centric, often depend on private data and expensive paired-report supervision, and do not explicitly model relationships among multiple slides from the same patient. We present MOOZY, a patient-first pathology foundation model in which the patient case, not the individual slide, is the core unit of representation. MOOZY explicitly models dependencies across all slides from the same patient via a case transformer during pretraining, combining multi-stage open self-supervision with scaled low-cost task supervision. In Stage 1, we pretrain a vision-only slide encoder on 77,134 public slide feature grids using masked self-distillation. In Stage 2, we align these representations with clinical semantics using a case transformer and multi-task supervision over 333 tasks from 56 public datasets, including 205 classification and 128 survival tasks across four endpoints. Across eight held-out tasks with five-fold frozen-feature probe evaluation, MOOZY achieves best or tied-best performance on most metrics and improves macro averages over TITAN by +7.37%, +5.50%, and +7.83% and over PRISM by +8.83%, +10.70%, and +9.78% for weighted F1, weighted ROC-AUC, and balanced accuracy, respectively. MOOZY is also parameter efficient with 85.77M parameters, 14x smaller than GigaPath. These results demonstrate that open, reproducible patient-level pretraining yields transferable embeddings, providing a practical path toward scalable patient-first histopathology foundation models.

Modeling spinal motion is fundamental to understanding human biomechanics, yet remains underexplored in computer vision due to the spine's complex multi-joint kinematics and the lack of large-scale 3D annotations. We present a biomechanics-aware keypoint simulation framework that augments existing human pose datasets with anatomically consistent 3D spinal keypoints derived from musculoskeletal modeling. Using this framework, we create the first open dataset, named SIMSPINE, which provides sparse vertebra-level 3D spinal annotations for natural full-body motions in indoor multi-camera capture without external restraints. With 2.14 million frames, this enables data-driven learning of vertebral kinematics from subtle posture variations and bridges the gap between musculoskeletal simulation and computer vision. In addition, we release pretrained baselines covering fine-tuned 2D detectors, monocular 3D pose lifting models, and multi-view reconstruction pipelines, establishing a unified benchmark for biomechanically valid spine motion estimation. Specifically, our 2D spine baselines improve the state-of-the-art from 0.63 to 0.80 AUC in controlled environments, and from 0.91 to 0.93 AP for in-the-wild spine tracking. Together, the simulation framework and SIMSPINE dataset advance research in vision-based biomechanics, motion analysis, and digital human modeling by enabling reproducible, anatomically grounded 3D spine estimation under natural conditions.

Recent advances in music foundation models have improved audio representation learning, yet their effectiveness across diverse musical traditions remains limited. We introduce CultureMERT-95M, a multi-culturally adapted foundation model developed to enhance cross-cultural music representation learning and understanding. To achieve this, we propose a two-stage continual pre-training strategy that integrates learning rate re-warming and re-decaying, enabling stable adaptation even with limited computational resources. Training on a 650-hour multi-cultural data mix, comprising Greek, Turkish, and Indian music traditions, results in an average improvement of 4.9% in ROC-AUC and AP across diverse non-Western music auto-tagging tasks, surpassing prior state-of-the-art, with minimal forgetting on Western-centric benchmarks. We further investigate task arithmetic, an alternative approach to multi-cultural adaptation that merges single-culture adapted models in the weight space. Task arithmetic performs on par with our multi-culturally trained model on non-Western auto-tagging tasks and shows no regression on Western datasets. Cross-cultural evaluation reveals that single-culture models transfer with varying effectiveness across musical traditions, whereas the multi-culturally adapted model achieves the best overall performance. To support research on world music representation learning, we publicly release CultureMERT-95M and CultureMERT-TA-95M, fostering the development of more culturally aware music foundation models.

Superconductivity allows electrical current to flow without any energy loss, and thus making solids superconducting is a grand goal of physics, material science, and electrical engineering. More than 16 Nobel Laureates have been awarded for their contribution to superconductivity research. Superconductors are valuable for sustainable development goals (SDGs), such as climate change mitigation, affordable and clean energy, industry, innovation and infrastructure, and so on. However, a unified physics theory explaining all superconductivity mechanism is still unknown. It is believed that superconductivity is microscopically due to not only molecular compositions but also the geometric crystal structure. Hence a new dataset, S2S, containing both crystal structures and superconducting critical temperature, is built upon SuperCon and Material Project. Based on this new dataset, we propose a novel model, S2SNet, which utilizes the attention mechanism for superconductivity prediction. To overcome the shortage of data, S2SNet is pre-trained on the whole Material Project dataset with Masked-Language Modeling (MLM). S2SNet makes a new state-of-the-art, with out-of-sample accuracy of 92% and Area Under Curve (AUC) of 0.92. To the best of our knowledge, S2SNet is the first work to predict superconductivity with only information of crystal structures. This work is beneficial to superconductivity discovery and further SDGs. Code and datasets are available in https://github.com/zjuKeLiu/S2SNet

As chain of thought (CoT) has become central to scaling reasoning capabilities in large language models (LLMs), it has also emerged as a promising tool for interpretability, suggesting the opportunity to understand model decisions through verbalized reasoning. However, the utility of CoT toward interpretability depends upon its faithfulness---whether the model's stated reasoning reflects the underlying decision process. We provide mechanistic evidence that instruction-tuned models often determine their answer before generating CoT. Training linear probes on residual stream activations at the last token before CoT, we can predict the model's final answer with >0.9 AUC on most tasks. We find that these directions are not only predictive, but also causal: steering activations along the probe direction often flips model answers, with flip rates substantially exceeding norm-matched orthogonal baselines across most model-dataset pairs. When steering induces incorrect answers, we observe two distinct failure modes: confabulation (fabricating false premises) and non-entailment (stating correct premises but drawing unsupported conclusions). While post-hoc reasoning may be instrumentally useful when the model has a correct pre-CoT belief, these failure modes suggest it can result in undesirable behaviors when reasoning from a false belief.

How different populations visually explore artworks bears on cognitive science and on accessibility design, yet most eye-tracking work in autism has used social scenes rather than art, and has analysed where the eyes land while ignoring when and in what order. We present a comparative free-viewing study across three groups, autistic adults (ASD), trained artists, and neurotypical observers, who each viewed 30 paintings for 15s. We introduce a directed, metric-grounded framework that compares groups along two complementary axes: a spatial axis, in which one group's fixation-density map predicts another's fixations under six saliency metrics (AUC-Judd, NSS, CC, SIM, KL, Information Gain); and a temporal axis, in which individual scanpaths are compared with MultiMatch, ScanMatch, a foveal-disc IoU score (FDISS), and dynamic time warping (DTW). Fixations are extracted uniformly for all groups with a dispersion-threshold algorithm. Three results converge. (i)Artists and neurotypicals are almost indistinguishable in both space (density-map correlation CC=0.96) and time (they form the most alignable scanpath pair), whereas ASD gaze diverges from both. (ii)ASD attention is dissociated: it matches artists' wide spatial exploration (dispersion, explored area) but carries a distinct temporal signature, shorter fixations, less dwell, and the most idiosyncratic (least self-consistent) scanpaths of any group. (iii)ASD gaze is not selectively artist-like on any metric; if anything it is marginally closer to neurotypical. Together these findings indicate that autistic viewing of art is a distinct, group-specific attentional profile in both space and time, and they motivate population-conditioned models of aesthetic attention. We release all analysis code and per-stimulus results.

Pancreatic cancer is projected to be the second-deadliest cancer by 2030, making early detection critical. Intraductal papillary mucinous neoplasms (IPMNs), key cancer precursors, present a clinical dilemma, as current guidelines struggle to stratify malignancy risk, leading to unnecessary surgeries or missed diagnoses. Here, we introduce Cyst-X, a multi-center MRI benchmark and a federated learning framework for IPMN malignancy-risk stratification. The dataset comprises 1,461 abdominal MRI scans from 764 patients at seven international centers, with three-tier malignancy labels anchored in histopathology or three-year imaging follow-up and expert pancreas segmentations. The pipeline couples the PanSegNet pancreas segmenter with a 3D DenseNet-121 classifier and a parallel radiomics predictor. On internal cross-validation, the deep learning classifier reached a mean area under the receiver operating characteristic curve (AUC) of 0.85 (95% confidence interval 0.84-0.86) on T2-weighted MRI for high-risk versus low- or no-risk discrimination, with the average precision rising from a prevalence baseline of 0.23 to 0.64. This performance was preserved (AUC 0.85, FedProx) when training was distributed across institutions without exchange of raw patient images. Benchmarked against three blinded radiologists on a 629-case reader subset evaluated under imaging-only conditions, the classifier matched or exceeded sensitivity at comparable specificity. To accelerate research in early pancreatic cancer detection, we publicly release the Cyst-X dataset, segmentation masks, and trained models as the first large-scale, multi-centre MRI resource for pancreatic cystic neoplasm analysis.

CCTV safety monitoring demands anomaly detectors combine reliable clip-level accuracy with predictable per-clip latency despite weak supervision. This work investigates compact vision-language models (VLMs) as practical detectors for this regime. A unified evaluation protocol standardizes preprocessing, prompting, dataset splits, metrics, and runtime settings to compare parameter-efficiently adapted compact VLMs against training-free VLM pipelines and weakly supervised baselines. Evaluation spans accuracy, precision, recall, F1, ROC-AUC, and average per-clip latency to jointly quantify detection quality and efficiency. With parameter-efficient adaptation, compact VLMs achieve performance on par with, and in several cases exceeding, established approaches while retaining competitive per-clip latency. Adaptation further reduces prompt sensitivity, producing more consistent behavior across prompt regimes under the shared protocol. These results show that parameter-efficient fine-tuning enables compact VLMs to serve as dependable clip-level anomaly detectors, yielding a favorable accuracy-efficiency trade-off within a transparent and consistent experimental setup.

Retinal imaging has emerged as a powerful, non-invasive modality for detecting and quantifying biomarkers of systemic diseases-ranging from diabetes and hypertension to Alzheimer's disease and cardiovascular disorders but current insights remain dispersed across platforms and specialties. Recent technological advances in optical coherence tomography (OCT/OCTA) and adaptive optics (AO) now deliver ultra-high-resolution scans (down to 5 {\mu}m ) with superior contrast and spatial integration, allowing early identification of microvascular abnormalities and neurodegenerative changes. At the same time, AI-driven and machine learning (ML) algorithms have revolutionized the analysis of large-scale retinal datasets, increasing sensitivity and specificity; for example, deep learning models achieve > 90 \% sensitivity for diabetic retinopathy and AUC = 0.89 for the prediction of cardiovascular risk from fundus photographs. The proliferation of mobile health technologies and telemedicine platforms further extends access, reduces costs, and facilitates community-based screening and longitudinal monitoring. Despite these breakthroughs, translation into routine practice is hindered by heterogeneous imaging protocols, limited external validation of AI models, and integration challenges within clinical workflows. In this review, we systematically synthesize the latest OCT/OCT and AO developments, AI/ML approaches, and mHealth/Tele-ophthalmology initiatives and quantify their diagnostic performance across disease domains. Finally, we propose a roadmap for multicenter protocol standardization, prospective validation trials, and seamless incorporation of retinal screening into primary and specialty care pathways-paving the way for precision prevention, early intervention, and ongoing treatment of life-threatening systemic diseases.

Accurate, continuous out-of-hospital electrocardiogram (ECG) parameter measurement is vital for real-time cardiac health monitoring and telemedicine. On-device computation of single-lead ECG parameters enables timely assessment without reliance on centralized data processing, advancing personalized, ubiquitous cardiac care-yet comprehensive validation across heterogeneous real-world populations remains limited. This study validated the on-device algorithm FeatureDB (https://github.com/PKUDigitalHealth/FeatureDB) using two datasets: HeartVoice-ECG-lite (369 participants with single-lead ECGs annotated by two physicians) and PTB-XL/PTB-XL+ (21,354 patients with 12-lead ECGs and physicians' diagnostic annotations). FeatureDB computed PR, QT, and QTc intervals, with accuracy evaluated against physician annotations via mean absolute error (MAE), correlation analysis, and Bland-Altman analysis. Diagnostic performance for first-degree atrioventricular block (AVBI, PR-based) and long QT syndrome (LQT, QTc-based) was benchmarked against commercial 12-lead systems (12SL, Uni-G) and open-source algorithm Deli, using AUC, accuracy, sensitivity, and specificity. Results showed high concordance with expert annotations (Pearson correlations: 0.836-0.960), MAEs matching inter-observer variability, and minimal bias. AVBI AUC reached 0.787 (12SL: 0.859; Uni-G: 0.812; Deli: 0.501); LQT AUC was 0.684 (12SL: 0.716; Uni-G: 0.605; Deli: 0.569)-comparable to commercial tools and superior to open-source alternatives. FeatureDB delivers physician-level parameter accuracy and commercial-grade abnormality detection via single-lead devices, supporting scalable telemedicine, decentralized cardiac screening, and continuous monitoring in community and outpatient settings.

Localizing the exact pathological regions in a given medical scan is an important imaging problem that traditionally requires a large amount of bounding box ground truth annotations to be accurately solved. However, there exist alternative, potentially weaker, forms of supervision, such as accompanying free-text reports, which are readily available. The task of performing localization with textual guidance is commonly referred to as phrase grounding. In this work, we use a publicly available Foundation Model, namely the Latent Diffusion Model, to perform this challenging task. This choice is supported by the fact that the Latent Diffusion Model, despite being generative in nature, contains cross-attention mechanisms that implicitly align visual and textual features, thus leading to intermediate representations that are suitable for the task at hand. In addition, we aim to perform this task in a zero-shot manner, i.e., without any training on the target task, meaning that the model's weights remain frozen. To this end, we devise strategies to select features and also refine them via post-processing without extra learnable parameters. We compare our proposed method with state-of-the-art approaches which explicitly enforce image-text alignment in a joint embedding space via contrastive learning. Results on a popular chest X-ray benchmark indicate that our method is competitive with SOTA on different types of pathology, and even outperforms them on average in terms of two metrics (mean IoU and AUC-ROC). Source code will be released upon acceptance at https://github.com/vios-s.

Computer-aided histopathological image analysis for cancer detection is a major research challenge in the medical domain. Automatic detection and classification of nuclei for cancer diagnosis impose a lot of challenges in developing state of the art algorithms due to the heterogeneity of cell nuclei and data set variability. Recently, a multitude of classification algorithms has used complex deep learning models for their dataset. However, most of these methods are rigid and their architectural arrangement suffers from inflexibility and non-interpretability. In this research article, we have proposed a hybrid and flexible deep learning architecture OLConvNet that integrates the interpretability of traditional object-level features and generalization of deep learning features by using a shallower Convolutional Neural Network (CNN) named as CNN_{3L}. CNN_{3L} reduces the training time by training fewer parameters and hence eliminating space constraints imposed by deeper algorithms. We used F1-score and multiclass Area Under the Curve (AUC) performance parameters to compare the results. To further strengthen the viability of our architectural approach, we tested our proposed methodology with state of the art deep learning architectures AlexNet, VGG16, VGG19, ResNet50, InceptionV3, and DenseNet121 as backbone networks. After a comprehensive analysis of classification results from all four architectures, we observed that our proposed model works well and perform better than contemporary complex algorithms.

Large pre-trained language models (LMs) such as GPT-3 have acquired a surprising ability to perform zero-shot learning. For example, to classify sentiment without any training examples, we can "prompt" the LM with the review and the label description "Does the user like this movie?", and ask whether the next word is "yes" or "no". However, the next word prediction training objective is still misaligned with the target zero-shot learning objective. To address this weakness, we propose meta-tuning, which directly optimizes the zero-shot learning objective by fine-tuning pre-trained language models on a collection of datasets. We focus on classification tasks, and construct the meta-dataset by aggregating 43 existing datasets and annotating 441 label descriptions in a question-answering (QA) format. When evaluated on unseen tasks, meta-tuned models outperform a same-sized QA model and the previous SOTA zero-shot learning system based on natural language inference. Additionally, increasing parameter count from 220M to 770M improves AUC-ROC scores by 6.3%, and we forecast that even larger models would perform better. Therefore, measuring zero-shot learning performance on language models out-of-the-box might underestimate their true potential, and community-wide efforts on aggregating datasets and unifying their formats can help build models that answer prompts better.

In-context learning (ICL) is the standard method for low-resource classification, yet its efficacy in specialized domains remains largely unexplored. We address the challenge of classifying semantically complex, multi-party B2B conversations, where traditional ICL encounters significant limitations, especially as context length increases due to the concatenation of multiple few-shot examples. We introduce the Call Playbook dataset, featuring five classification tasks derived from real-world B2B conversations targeting core sales concepts. To bridge the gap between performance and practical utility, we propose novel knowledge extraction methods that distill verbose examples into compact, interpretable representations of structured classification criteria and precise task descriptions. Our approach achieves a 99\% reduction in token usage and improves macro-averaged AUC by up to 7\% over traditional ICL. Notably, it remains robust as context grows, unlike advanced token compression baselines which degrade by over 9 F1 points. Importantly, our framework enables direct refinement of classification logic, addressing critical needs for transparency, efficiency, and user interaction in real-world NLP applications.

Large vision-language models (LVLMs) have been regarded as a breakthrough advance in an astoundingly variety of tasks, from content generation to virtual assistants and multimodal search or retrieval. However, for many of these applications, the performance of these methods has been widely criticized, particularly when compared with state-of-the-art methods and technologies in each specific domain. In this work, we compare the performance of the leading large vision-language models in the human re-identification task, using as baseline the performance attained by state-of-the-art AI models specifically designed for this problem. We compare the results due to ChatGPT-4o, Gemini-2.0-Flash, Claude 3.5 Sonnet, and Qwen-VL-Max to a baseline ReID PersonViT model, using the well-known Market1501 dataset. Our evaluation pipeline includes the dataset curation, prompt engineering, and metric selection to assess the models' performance. Results are analyzed from many different perspectives: similarity scores, classification accuracy, and classification metrics, including precision, recall, F1 score, and area under curve (AUC). Our results confirm the strengths of LVLMs, but also their severe limitations that often lead to catastrophic answers and should be the scope of further research. As a concluding remark, we speculate about some further research that should fuse traditional and LVLMs to combine the strengths from both families of techniques and achieve solid improvements in performance.

When a language model asserts that "the capital of Australia is Sydney," does it know this is wrong? We characterize the geometry of correctness representations across 9 models from 5 architecture families. The structure is simple: the discriminative signal occupies 3-8 dimensions, performance degrades with additional dimensions, and no nonlinear classifier improves over linear separation. Centroid distance in the low-dimensional subspace matches trained probe performance (0.90 AUC), enabling few-shot detection: on GPT-2, 25 labeled examples achieve 89% of full-data accuracy. We validate causally through activation steering: the learned direction produces 10.9 percentage point changes in error rates while random directions show no effect. Internal probes achieve 0.80-0.97 AUC; output-based methods (P(True), semantic entropy) achieve only 0.44-0.64 AUC. The correctness signal exists internally but is not expressed in outputs. That centroid distance matches probe performance indicates class separation is a mean shift, making detection geometric rather than learned.

Video Anomaly Detection (VAD) aims to identify anomalous events within video segments. In scenarios such as surveillance or industrial process monitoring, anomaly detection is of critical importance. While existing approaches are semi-automated, requiring human assessment for anomaly detection, traditional VADs offer limited output as either normal or anomalous. We propose Text-guided Fine-Grained Video Anomaly Detection (T-VAD), a framework built upon Large Vision-Language Model (LVLM). T-VAD introduces an Anomaly Heatmap Decoder (AHD) that performs pixel-wise visual-textual feature alignment to generate fine-grained anomaly heatmaps. Furthermore, we design a Region-aware Anomaly Encoder (RAE) that transforms the heatmaps into learnable textual embeddings, guiding the LVLM to accurately identify and localize anomalous events in videos. This significantly enhances both the granularity and interactivity of anomaly detection. The proposed method achieving SOTA performance by demonstrating 94.8% Area Under the Curve (AUC, specifically micro-AUC) and 67.8%/76.7% accuracy in anomaly heatmaps (RBDC/TBDC) on the UBnormal dataset, and subjectively verified more preferable textual description on the ShanghaiTech-based dataset (BLEU-4: 62.67 for targets, 88.84 for trajectories; Yes/No accuracy: 97.67%), and on the UBnormal dataset (BLEU-4: 50.32 for targets, 78.10 for trajectories; Yes/No accuracy: 89.73%).

Some recent studies have described deep convolutional neural networks to diagnose breast cancer in mammograms with similar or even superior performance to that of human experts. One of the best techniques does two transfer learnings: the first uses a model trained on natural images to create a "patch classifier" that categorizes small subimages; the second uses the patch classifier to scan the whole mammogram and create the "single-view whole-image classifier". We propose to make a third transfer learning to obtain a "two-view classifier" to use the two mammographic views: bilateral craniocaudal and mediolateral oblique. We use EfficientNet as the basis of our model. We "end-to-end" train the entire system using CBIS-DDSM dataset. To ensure statistical robustness, we test our system twice using: (a) 5-fold cross validation; and (b) the original training/test division of the dataset. Our technique reached an AUC of 0.9344 using 5-fold cross validation (accuracy, sensitivity and specificity are 85.13% at the equal error rate point of ROC). Using the original dataset division, our technique achieved an AUC of 0.8483, as far as we know the highest reported AUC for this problem, although the subtle differences in the testing conditions of each work do not allow for an accurate comparison. The inference code and model are available at https://github.com/dpetrini/two-views-classifier

Despite being (pre)trained on a massive amount of data, state-of-the-art video-language alignment models are not robust to semantically-plausible contrastive changes in the video captions. Our work addresses this by identifying a broad spectrum of contrast misalignments, such as replacing entities, actions, and flipping event order, which alignment models should be robust against. To this end, we introduce the VideoCon, a video-language alignment dataset constructed by a large language model that generates plausible contrast video captions and explanations for differences between original and contrast video captions. Then, a generative video-language model is finetuned with VideoCon to assess video-language entailment and generate explanations. Our VideoCon-based alignment model significantly outperforms current models. It exhibits a 12-point increase in AUC for the video-language alignment task on human-generated contrast captions. Finally, our model sets new state of the art zero-shot performance in temporally-extensive video-language tasks such as text-to-video retrieval (SSv2-Temporal) and video question answering (ATP-Hard). Moreover, our model shows superior performance on novel videos and human-crafted captions and explanations. Our code and data are available at https://github.com/Hritikbansal/videocon.

Continuous glucose monitoring (CGM) provides a dense view of daily metabolic physiology, yet existing generic time-series and CGM-specific foundation models often encode glucose traces as entangled single-stream sequences, leaving the distinct temporal structure of glycemic dynamics only implicitly modeled. We present GlucoFM, a lightweight CGM foundation model that aligns irregular recordings to a 24-hour chronological grid, preserves observation masks, and decomposes glucose dynamics into slow physiological state and transient event streams, capturing low-frequency glycemic baselines and short-term deviations that may reflect acute physiological responses or sensor artifacts. GlucoFM is pretrained on 109,066 hours of unlabeled CGM recordings from 477 subjects with two complementary objectives: masked contextual latent prediction over fused daily representations and temporal dynamics prediction over state and event streams. Across four diverse cohorts and seven clinical prediction tasks, GlucoFM achieves the strongest subject-disjoint linear-probing performance among evaluated baselines, improving average PR-AUC by 4.1 points over the best CGM-specific foundation model. Its gains are most pronounced on core metabolic outcomes, leading PR-AUC on all diabetes-risk and β-cell dysfunction tasks and on 3 of 4 insulin-resistance tasks. GlucoFM also achieves the best overall cross-dataset transfer performance and strong few-shot adaptation among evaluated methods, and consistent gains when aggregating multiple days for subject-level prediction, highlighting physiology-aware decomposition as an effective inductive bias for transferable CGM representation learning.

Most image watermarking systems focus on robustness, capacity, and imperceptibility while treating the embedded payload as meaningless bits. This bit-centric view imposes a hard ceiling on capacity and prevents watermarks from carrying useful information. We propose LatentSeal, which reframes watermarking as semantic communication: a lightweight text autoencoder maps full-sentence messages into a compact 256-dimensional unit-norm latent vector, which is robustly embedded by a finetuned watermark model and secured through a secret, invertible rotation. The resulting system hides full-sentence messages, decodes in real time, and survives valuemetric and geometric attacks. It surpasses prior state of the art in BLEU-4 and Exact Match on several benchmarks, while breaking through the long-standing 256-bit payload ceiling. It also introduces a statistically calibrated score that yields a ROC AUC score of 0.97-0.99, and practical operating points for deployment. By shifting from bit payloads to semantic latent vectors, LatentSeal enables watermarking that is not only robust and high-capacity, but also secure and interpretable, providing a concrete path toward provenance, tamper explanation, and trustworthy AI governance. Models, training and inference code, and data splits will be available upon publication.

Accurate localization of tumor regions from hematoxylin and eosin-stained whole-slide images is fundamental for translational research including spatial analysis, molecular profiling, and tissue architecture investigation. However, deep learning-based tumor detection trained within specific cancers may exhibit reduced robustness when applied across different tumor types. We investigated whether balanced training across cancers at modest scale can achieve high performance and generalize to unseen tumor types. A multi-cancer tumor localization model (MuCTaL) was trained on 79,984 non-overlapping tiles from four cancers (melanoma, hepatocellular carcinoma, colorectal cancer, and non-small cell lung cancer) using transfer learning with DenseNet169. The model achieved a tile-level ROC-AUC of 0.97 in validation data from the four training cancers, and 0.71 on an independent pancreatic ductal adenocarcinoma cohort. A scalable inference workflow was built to generate spatial tumor probability heatmaps compatible with existing digital pathology tools. Code and models are publicly available at https://github.com/AivaraX-AI/MuCTaL.

Recent 3D CT vision-language models align volumes with reports via contrastive pretraining, but typically rely on limited public data and provide only coarse global supervision. We train a 3D CT vision-language model on 98k report-volume pairs (50k patients) collected at a single hospital, combined with public datasets, using SigLIP-style contrastive pretraining together with prompt-based disease supervision in the shared vision-text embedding space. On CT-RATE, our model achieves state-of-the-art text-to-image retrieval (R@10 31.5 vs. 22.2) and competitive disease classification (AUC 83.8 vs. 83.8), with consistent results on Rad-ChestCT (AUC 77.0 vs. 77.3). We further observe that radiologists routinely reference specific images within their reports (e.g., ``series X, image Y''), linking textual descriptions to precise axial locations. We automatically mine 262k such snippet-slice pairs and introduce the task of intra-scan snippet localization -- predicting the axial depth referred to by a text snippet -- reducing mean absolute error to 36.3 mm at 12 mm feature resolution, compared with 67.0 mm for the best baseline. Adding this localization objective leaves retrieval and classification broadly unchanged within confidence bounds, yielding a single unified model for retrieval, classification, and intra-scan grounding.

Reasoning distillation has emerged as an efficient and powerful paradigm for enhancing the reasoning capabilities of large language models. However, reasoning distillation may inadvertently cause benchmark contamination, where evaluation data included in distillation datasets can inflate performance metrics of distilled models. In this work, we formally define the task of distillation data detection, which is uniquely challenging due to the partial availability of distillation data. Then, we propose a novel and effective method Token Probability Deviation (TBD), which leverages the probability patterns of the generated output tokens. Our method is motivated by the analysis that distilled models tend to generate near-deterministic tokens for seen questions, while producing more low-probability tokens for unseen questions. Our key idea behind TBD is to quantify how far the generated tokens' probabilities deviate from a high reference probability. In effect, our method achieves competitive detection performance by producing lower scores for seen questions than for unseen questions. Extensive experiments demonstrate the effectiveness of our method, achieving an AUC of 0.918 and a TPR@1% FPR of 0.470 on the S1 dataset.

We present RAVEN, a newly developed vetting and validation pipeline for TESS exoplanet candidates. The pipeline employs a Bayesian framework to derive the posterior probability of a candidate being a planet against a set of False Positive (FP) scenarios, through the use of a Gradient Boosted Decision Tree and a Gaussian Process classifier, trained on comprehensive synthetic training sets of simulated planets and 8 astrophysical FP scenarios injected into TESS lightcurves. These training sets allow large scale candidate vetting and performance verification against individual FP scenarios. A Non-Simulated FP training set consisting of real TESS candidates caused primarily by stellar variability and systematic noise is also included. The machine learning derived probabilities are combined with scenario specific prior probabilities, including the candidates' positional probabilities, to compute the final posterior probabilities. Candidates with a planetary posterior probability greater than 99% against each FP scenario and whose implied planetary radius is less than 8R_{oplus} are considered to be statistically validated by the pipeline. In this first version, the pipeline has been developed for candidates with a lightcurve released from the TESS Science Processing Operations Centre, an orbital period between 0.5 and 16 days and a transit depth greater than 300ppm. The pipeline obtained area-under-curve (AUC) scores > 97% on all FP scenarios and > 99% on all but one. Testing on an independent external sample of 1361 pre-classified TOIs, the pipeline achieved an overall accuracy of 91%, demonstrating its effectiveness for automated ranking of TESS candidates. For a probability threshold of 0.9 the pipeline reached a precision of 97% with a recall score of 66% on these TOIs. The RAVEN pipeline is publicly released as a cloud-hosted app, making it easily accessible to the community.

Accurate differentiation of pseudoprogression (PsP) from True Progression (TP) following radiotherapy (RT) in glioblastoma (GBM) patients is crucial for optimal treatment planning. However, this task remains challenging due to the overlapping imaging characteristics of PsP and TP. This study therefore proposes a multimodal deep-learning approach utilizing complementary information from routine anatomical MR images, clinical parameters, and RT treatment planning information for improved predictive accuracy. The approach utilizes a self-supervised Vision Transformer (ViT) to encode multi-sequence MR brain volumes to effectively capture both global and local context from the high dimensional input. The encoder is trained in a self-supervised upstream task on unlabeled glioma MRI datasets from the open BraTS2021, UPenn-GBM, and UCSF-PDGM datasets to generate compact, clinically relevant representations from FLAIR and T1 post-contrast sequences. These encoded MR inputs are then integrated with clinical data and RT treatment planning information through guided cross-modal attention, improving progression classification accuracy. This work was developed using two datasets from different centers: the Burdenko Glioblastoma Progression Dataset (n = 59) for training and validation, and the GlioCMV progression dataset from the University Hospital Erlangen (UKER) (n = 20) for testing. The proposed method achieved an AUC of 75.3%, outperforming the current state-of-the-art data-driven approaches. Importantly, the proposed approach relies on readily available anatomical MRI sequences, clinical data, and RT treatment planning information, enhancing its clinical feasibility. The proposed approach addresses the challenge of limited data availability for PsP and TP differentiation and could allow for improved clinical decision-making and optimized treatment plans for GBM patients.

We present a method based on natural language processing (NLP), for studying the influence of interest groups (lobbies) in the law-making process in the European Parliament (EP). We collect and analyze novel datasets of lobbies' position papers and speeches made by members of the EP (MEPs). By comparing these texts on the basis of semantic similarity and entailment, we are able to discover interpretable links between MEPs and lobbies. In the absence of a ground-truth dataset of such links, we perform an indirect validation by comparing the discovered links with a dataset, which we curate, of retweet links between MEPs and lobbies, and with the publicly disclosed meetings of MEPs. Our best method achieves an AUC score of 0.77 and performs significantly better than several baselines. Moreover, an aggregate analysis of the discovered links, between groups of related lobbies and political groups of MEPs, correspond to the expectations from the ideology of the groups (e.g., center-left groups are associated with social causes). We believe that this work, which encompasses the methodology, datasets, and results, is a step towards enhancing the transparency of the intricate decision-making processes within democratic institutions.

The Multimodal Sentiment Analysis Challenge (MuSe) 2022 is dedicated to multimodal sentiment and emotion recognition. For this year's challenge, we feature three datasets: (i) the Passau Spontaneous Football Coach Humor (Passau-SFCH) dataset that contains audio-visual recordings of German football coaches, labelled for the presence of humour; (ii) the Hume-Reaction dataset in which reactions of individuals to emotional stimuli have been annotated with respect to seven emotional expression intensities, and (iii) the Ulm-Trier Social Stress Test (Ulm-TSST) dataset comprising of audio-visual data labelled with continuous emotion values (arousal and valence) of people in stressful dispositions. Using the introduced datasets, MuSe 2022 2022 addresses three contemporary affective computing problems: in the Humor Detection Sub-Challenge (MuSe-Humor), spontaneous humour has to be recognised; in the Emotional Reactions Sub-Challenge (MuSe-Reaction), seven fine-grained `in-the-wild' emotions have to be predicted; and in the Emotional Stress Sub-Challenge (MuSe-Stress), a continuous prediction of stressed emotion values is featured. The challenge is designed to attract different research communities, encouraging a fusion of their disciplines. Mainly, MuSe 2022 targets the communities of audio-visual emotion recognition, health informatics, and symbolic sentiment analysis. This baseline paper describes the datasets as well as the feature sets extracted from them. A recurrent neural network with LSTM cells is used to set competitive baseline results on the test partitions for each sub-challenge. We report an Area Under the Curve (AUC) of .8480 for MuSe-Humor; .2801 mean (from 7-classes) Pearson's Correlations Coefficient for MuSe-Reaction, as well as .4931 Concordance Correlation Coefficient (CCC) and .4761 for valence and arousal in MuSe-Stress, respectively.

Moderators and automated methods enforce bans on malicious users who engage in disruptive behavior. However, malicious users can easily create a new account to evade such bans. Previous research has focused on other forms of online deception, like the simultaneous operation of multiple accounts by the same entities (sockpuppetry), impersonation of other individuals, and studying the effects of de-platforming individuals and communities. Here we conduct the first data-driven study of ban evasion, i.e., the act of circumventing bans on an online platform, leading to temporally disjoint operation of accounts by the same user. We curate a novel dataset of 8,551 ban evasion pairs (parent, child) identified on Wikipedia and contrast their behavior with benign users and non-evading malicious users. We find that evasion child accounts demonstrate similarities with respect to their banned parent accounts on several behavioral axes - from similarity in usernames and edited pages to similarity in content added to the platform and its psycholinguistic attributes. We reveal key behavioral attributes of accounts that are likely to evade bans. Based on the insights from the analyses, we train logistic regression classifiers to detect and predict ban evasion at three different points in the ban evasion lifecycle. Results demonstrate the effectiveness of our methods in predicting future evaders (AUC = 0.78), early detection of ban evasion (AUC = 0.85), and matching child accounts with parent accounts (MRR = 0.97). Our work can aid moderators by reducing their workload and identifying evasion pairs faster and more efficiently than current manual and heuristic-based approaches. Dataset is available https://github.com/srijankr/ban_evasion{here}.

Electronic health record foundation models are limited by institutionally siloed data and substantial performance degradation under cross-site transfer. We evaluated federated training of tokenized generative event models (GEMs) across 122,251 intensive care hospitalizations from three independent health systems harmonized to the Common Longitudinal ICU Data Format. Models were assessed on 12 post-24-hour clinical prediction tasks using within-site, cross-site, centralized, and federated training configurations. GEMs achieved the highest mean within-site and cross-site ROC-AUC and were substantially more transportable than conventional supervised models: their average cross-site penalties were 0.025 ROC-AUC and 0.027 PR-AUC, compared with 0.079 and 0.089 for LightGBM. Federated Learning (FedAvg and FedAvgM) approached the performance of centralized GEM training, with most gains obtained within 5-10 communication rounds. However, centralized multi-site training provided only modest improvements over complete local training. Multi-site models were most useful when local training data were limited, with their advantage narrowing as institutional data accumulated. These findings show that federated GEM training is technically feasible and preserves most centralized performance, but that the main open challenge is learning transportable representations to translate larger, but heterogeneous data from multiple health systems into a reliable target-site benefit.

Hospitals struggle to predict critical outcomes. Traditional early warning systems, like NEWS and MEWS, rely on static variables and fixed thresholds, limiting their adaptability, accuracy, and personalization. We previously developed the Enhanced Transformer for Health Outcome Simulation (ETHOS), an AI model that tokenizes patient health timelines (PHTs) from EHRs and uses transformer-based architectures to predict future PHTs. ETHOS is a versatile framework for developing a wide range of applications. In this work, we develop the Adaptive Risk Estimation System (ARES) that leverages ETHOS to compute dynamic, personalized risk probabilities for clinician-defined critical events. ARES also features a personalized explainability module that highlights key clinical factors influencing risk estimates. We evaluated ARES using the MIMIC-IV v2.2 dataset together with its Emergency Department (ED) extension and benchmarked performance against both classical early warning systems and contemporary machine learning models. The entire dataset was tokenized resulting in 285,622 PHTs, comprising over 360 million tokens. ETHOS outperformed benchmark models in predicting hospital admissions, ICU admissions, and prolonged stays, achieving superior AUC scores. Its risk estimates were robust across demographic subgroups, with calibration curves confirming model reliability. The explainability module provided valuable insights into patient-specific risk factors. ARES, powered by ETHOS, advances predictive healthcare AI by delivering dynamic, real-time, personalized risk estimation with patient-specific explainability. Although our results are promising, the clinical impact remains uncertain. Demonstrating ARES's true utility in real-world settings will be the focus of our future work. We release the source code to facilitate future research.

Microscopic assessment of histopathology images is vital for accurate cancer diagnosis and treatment. Whole Slide Image (WSI) classification and captioning have become crucial tasks in computer-aided pathology. However, microscopic WSI face challenges such as redundant patches and unknown patch positions due to subjective pathologist captures. Moreover, generating automatic pathology captions remains a significant challenge. To address these issues, we introduce a novel GNN-ViTCap framework for classification and caption generation from histopathological microscopic images. First, a visual feature extractor generates patch embeddings. Redundant patches are then removed by dynamically clustering these embeddings using deep embedded clustering and selecting representative patches via a scalar dot attention mechanism. We build a graph by connecting each node to its nearest neighbors in the similarity matrix and apply a graph neural network to capture both local and global context. The aggregated image embeddings are projected into the language model's input space through a linear layer and combined with caption tokens to fine-tune a large language model. We validate our method on the BreakHis and PatchGastric datasets. GNN-ViTCap achieves an F1 score of 0.934 and an AUC of 0.963 for classification, along with a BLEU-4 score of 0.811 and a METEOR score of 0.569 for captioning. Experimental results demonstrate that GNN-ViTCap outperforms state of the art approaches, offering a reliable and efficient solution for microscopy based patient diagnosis.

Video Anomaly Detection (VAD) automatically identifies anomalous events from video, mitigating the need for human operators in large-scale surveillance deployments. However, three fundamental obstacles hinder real-world adoption: domain dependency and real-time constraints -- requiring near-instantaneous processing of incoming video. To this end, we propose Flashback, a zero-shot and real-time video anomaly detection paradigm. Inspired by the human cognitive mechanism of instantly judging anomalies and reasoning in current scenes based on past experience, Flashback operates in two stages: Recall and Respond. In the offline recall stage, an off-the-shelf LLM builds a pseudo-scene memory of both normal and anomalous captions without any reliance on real anomaly data. In the online respond stage, incoming video segments are embedded and matched against this memory via similarity search. By eliminating all LLM calls at inference time, Flashback delivers real-time VAD even on a consumer-grade GPU. On two large datasets from real-world surveillance scenarios, UCF-Crime and XD-Violence, we achieve 87.3 AUC (+7.0 pp) and 75.1 AP (+13.1 pp), respectively, outperforming prior zero-shot VAD methods by large margins.

This research introduces a novel AI techniques as Mixture-of-Experts Transformers with Group Relative Policy Optimization (GRPO) for voice health care applications on voice pathology detection. With the architectural innovations, we adopt advanced training paradigms inspired by reinforcement learning, namely Proximal Policy Optimization (PPO) and Group-wise Regularized Policy Optimization (GRPO), to enhance model stability and performance. Experiments conducted on a synthetically generated voice pathology dataset demonstrate that our proposed models significantly improve diagnostic accuracy, F1 score, and ROC-AUC compared to conventional approaches. These findings underscore the potential of integrating transformer architectures with novel training strategies to advance automated voice pathology detection and ultimately contribute to more effective healthcare delivery. The code we used to train and evaluate our models is available at https://github.com/enkhtogtokh/voicegrpo

Gathering histopathology slides from over 100 publicly available cohorts, we compile a diverse dataset of 460 million pathology tiles covering more than 30 cancer sites. Using this dataset, we train a large self-supervised vision transformer using DINOv2 and publicly release one iteration of this model for further experimentation, coined Phikon-v2. While trained on publicly available histology slides, Phikon-v2 surpasses our previously released model (Phikon) and performs on par with other histopathology foundation models (FM) trained on proprietary data. Our benchmarks include eight slide-level tasks with results reported on external validation cohorts avoiding any data contamination between pre-training and evaluation datasets. Our downstream training procedure follows a simple yet robust ensembling strategy yielding a +1.75 AUC increase across tasks and models compared to one-shot retraining (p<0.001). We compare Phikon (ViT-B) and Phikon-v2 (ViT-L) against 14 different histology feature extractors, making our evaluation the most comprehensive to date. Our result support evidences that DINOv2 handles joint model and data scaling better than iBOT. Also, we show that recent scaling efforts are overall beneficial to downstream performance in the context of biomarker prediction with GigaPath and H-Optimus-0 (two ViT-g with 1.1B parameters each) standing out. However, the statistical margins between the latest top-performing FMs remain mostly non-significant; some even underperform on specific indications or tasks such as MSI prediction - deposed by a 13x smaller model developed internally. While latest foundation models may exhibit limitations for clinical deployment, they nonetheless offer excellent grounds for the development of more specialized and cost-efficient histology encoders fueling AI-guided diagnostic tools.

We present a Multimodal Backdoor Defense technique TIJO (Trigger Inversion using Joint Optimization). Recent work arXiv:2112.07668 has demonstrated successful backdoor attacks on multimodal models for the Visual Question Answering task. Their dual-key backdoor trigger is split across two modalities (image and text), such that the backdoor is activated if and only if the trigger is present in both modalities. We propose TIJO that defends against dual-key attacks through a joint optimization that reverse-engineers the trigger in both the image and text modalities. This joint optimization is challenging in multimodal models due to the disconnected nature of the visual pipeline which consists of an offline feature extractor, whose output is then fused with the text using a fusion module. The key insight enabling the joint optimization in TIJO is that the trigger inversion needs to be carried out in the object detection box feature space as opposed to the pixel space. We demonstrate the effectiveness of our method on the TrojVQA benchmark, where TIJO improves upon the state-of-the-art unimodal methods from an AUC of 0.6 to 0.92 on multimodal dual-key backdoors. Furthermore, our method also improves upon the unimodal baselines on unimodal backdoors. We present ablation studies and qualitative results to provide insights into our algorithm such as the critical importance of overlaying the inverted feature triggers on all visual features during trigger inversion. The prototype implementation of TIJO is available at https://github.com/SRI-CSL/TIJO.

Sensitivity to staining variation remains a major barrier to deploying computational pathology (CPath) models as hematoxylin and eosin (H&E) staining varies across laboratories, requiring systematic assessment of how this variability affects model prediction. In this work, we developed a three-step protocol for evaluating robustness to H&E staining variation in CPath models. Step 1: Select reference staining conditions, Step 2: Characterize test set staining properties, Step 3: Apply CPath model(s) under simulated reference staining conditions. Here, we first created a new reference staining library based on the PLISM dataset. As an exemplary use case, we applied the protocol to assess the robustness properties of 306 microsatellite instability (MSI) classification models on the unseen SurGen colorectal cancer dataset (n=738), including 300 attention-based multiple instance learning models trained on the TCGA-COAD/READ datasets across three feature extractors (UNI2-h, H-Optimus-1, Virchow2), alongside six public MSI classification models. Classification performance was measured as AUC, and robustness as the min-max AUC range across four simulated staining conditions (low/high H&E intensity, low/high H&E color similarity). Across models and staining conditions, classification performance ranged from AUC 0.769-0.911 (Δ = 0.142). Robustness ranged from 0.007-0.079 (Δ = 0.072), and showed a weak inverse correlation with classification performance (Pearson r=-0.22, 95% CI [-0.34, -0.11]). Thus, we show that the proposed evaluation protocol enables robustness-informed CPath model selection and provides insight into performance shifts across H&E staining conditions, supporting the identification of operational ranges for reliable model deployment. Code is available at https://github.com/CTPLab/staining-robustness-evaluation .

In this study, we introduce a novel method to predict mental health by building machine learning models for a non-invasive wearable device equipped with Laser Doppler Flowmetry (LDF) and Fluorescence Spectroscopy (FS) sensors. Besides, we present the corresponding dataset to predict mental health, e.g. depression, anxiety, and stress levels via the DAS-21 questionnaire. To our best knowledge, this is the world's largest and the most generalized dataset ever collected for both LDF and FS studies. The device captures cutaneous blood microcirculation parameters, and wavelet analysis of the LDF signal extracts key rhythmic oscillations. The dataset, collected from 132 volunteers aged 18-94 from 19 countries, explores relationships between physiological features, demographics, lifestyle habits, and health conditions. We employed a variety of machine learning methods to classify stress detection, in which LightGBM is identified as the most effective model for stress detection, achieving a ROC AUC of 0.7168 and a PR AUC of 0.8852. In addition, we also incorporated Explainable Artificial Intelligence (XAI) techniques into our analysis to investigate deeper insights into the model's predictions. Our results suggest that females, younger individuals and those with a higher Body Mass Index (BMI) or heart rate have a greater likelihood of experiencing mental health conditions like stress and anxiety. All related code and data are published online: https://github.com/leduckhai/Wearable_LDF-FS.

Developing robust deep learning models for fetal ultrasound image analysis requires comprehensive, high-quality datasets to effectively learn informative data representations within the domain. However, the scarcity of labelled ultrasound images poses substantial challenges, especially in low-resource settings. To tackle this challenge, we leverage synthetic data to enhance the generalizability of deep learning models. This study proposes a diffusion-based method, Fetal Ultrasound LoRA (FU-LoRA), which involves fine-tuning latent diffusion models using the LoRA technique to generate synthetic fetal ultrasound images. These synthetic images are integrated into a hybrid dataset that combines real-world and synthetic images to improve the performance of zero-shot classifiers in low-resource settings. Our experimental results on fetal ultrasound images from African cohorts demonstrate that FU-LoRA outperforms the baseline method by a 13.73% increase in zero-shot classification accuracy. Furthermore, FU-LoRA achieves the highest accuracy of 82.40%, the highest F-score of 86.54%, and the highest AUC of 89.78%. It demonstrates that the FU-LoRA method is effective in the zero-shot classification of fetal ultrasound images in low-resource settings. Our code and data are publicly accessible on https://github.com/13204942/FU-LoRA.

It is often difficult to correctly infer a writer's emotion from text exchanged online, and differences in recognition between writers and readers can be problematic. In this paper, we propose a new framework for detecting sentences that create differences in emotion recognition between the writer and the reader and for detecting the kinds of expressions that cause such differences. The proposed framework consists of a bidirectional encoder representations from transformers (BERT)-based detector that detects sentences causing differences in emotion recognition and an analysis that acquires expressions that characteristically appear in such sentences. The detector, based on a Japanese SNS-document dataset with emotion labels annotated by both the writer and three readers of the social networking service (SNS) documents, detected "hidden-anger sentences" with AUC = 0.772; these sentences gave rise to differences in the recognition of anger. Because SNS documents contain many sentences whose meaning is extremely difficult to interpret, by analyzing the sentences detected by this detector, we obtained several expressions that appear characteristically in hidden-anger sentences. The detected sentences and expressions do not convey anger explicitly, and it is difficult to infer the writer's anger, but if the implicit anger is pointed out, it becomes possible to guess why the writer is angry. Put into practical use, this framework would likely have the ability to mitigate problems based on misunderstandings.

Clinical trials are crucial for drug development but are time consuming, expensive, and often burdensome on patients. More importantly, clinical trials face uncertain outcomes due to issues with efficacy, safety, or problems with patient recruitment. If we were better at predicting the results of clinical trials, we could avoid having to run trials that will inevitably fail more resources could be devoted to trials that are likely to succeed. In this paper, we propose Hierarchical INteraction Network (HINT) for more general, clinical trial outcome predictions for all diseases based on a comprehensive and diverse set of web data including molecule information of the drugs, target disease information, trial protocol and biomedical knowledge. HINT first encode these multi-modal data into latent embeddings, where an imputation module is designed to handle missing data. Next, these embeddings will be fed into the knowledge embedding module to generate knowledge embeddings that are pretrained using external knowledge on pharmaco-kinetic properties and trial risk from the web. Then the interaction graph module will connect all the embedding via domain knowledge to fully capture various trial components and their complex relations as well as their influences on trial outcomes. Finally, HINT learns a dynamic attentive graph neural network to predict trial outcome. Comprehensive experimental results show that HINT achieves strong predictive performance, obtaining 0.772, 0.607, 0.623, 0.703 on PR-AUC for Phase I, II, III, and indication outcome prediction, respectively. It also consistently outperforms the best baseline method by up to 12.4\% on PR-AUC.

We propose TabTransformer, a novel deep tabular data modeling architecture for supervised and semi-supervised learning. The TabTransformer is built upon self-attention based Transformers. The Transformer layers transform the embeddings of categorical features into robust contextual embeddings to achieve higher prediction accuracy. Through extensive experiments on fifteen publicly available datasets, we show that the TabTransformer outperforms the state-of-the-art deep learning methods for tabular data by at least 1.0% on mean AUC, and matches the performance of tree-based ensemble models. Furthermore, we demonstrate that the contextual embeddings learned from TabTransformer are highly robust against both missing and noisy data features, and provide better interpretability. Lastly, for the semi-supervised setting we develop an unsupervised pre-training procedure to learn data-driven contextual embeddings, resulting in an average 2.1% AUC lift over the state-of-the-art methods.

Bird sounds possess distinctive spectral structure which may exhibit small shifts in spectrum depending on the bird species and environmental conditions. In this paper, we propose using convolutional recurrent neural networks on the task of automated bird audio detection in real-life environments. In the proposed method, convolutional layers extract high dimensional, local frequency shift invariant features, while recurrent layers capture longer term dependencies between the features extracted from short time frames. This method achieves 88.5% Area Under ROC Curve (AUC) score on the unseen evaluation data and obtains the second place in the Bird Audio Detection challenge.