Research Papers

Accurate diagnosis of brain disorders such as Alzheimer's disease and brain tumors remains a critical challenge in medical imaging. Conventional methods based on manual MRI analysis are often inefficient and error-prone. To address this, we propose DGG-XNet, a hybrid deep learning model integrating VGG16 and DenseNet121 to enhance feature extraction and classification. DenseNet121 promotes feature reuse and efficient gradient flow through dense connectivity, while VGG16 contributes strong hierarchical spatial representations. Their fusion enables robust multiclass classification of neurological conditions. Grad-CAM is applied to visualize salient regions, enhancing model transparency. Trained on a combined dataset from BraTS 2021 and Kaggle, DGG-XNet achieved a test accuracy of 91.33\%, with precision, recall, and F1-score all exceeding 91\%. These results highlight DGG-XNet's potential as an effective and interpretable tool for computer-aided diagnosis (CAD) of neurodegenerative and oncological brain disorders.

Existing end-to-end sign-language animation systems suffer from low naturalness, limited facial/body expressivity, and no user control. We propose a human-centered, real-time speech-to-sign animation framework that integrates (1) a streaming Conformer encoder with an autoregressive Transformer-MDN decoder for synchronized upper-body and facial motion generation, (2) a transparent, editable JSON intermediate representation empowering deaf users and experts to inspect and modify each sign segment, and (3) a human-in-the-loop optimization loop that refines the model based on user edits and ratings. Deployed on Unity3D, our system achieves a 13 ms average frame-inference time and a 103 ms end-to-end latency on an RTX 4070. Our key contributions include the design of a JSON-centric editing mechanism for fine-grained sign-level personalization and the first application of an MDN-based feedback loop for continuous model adaptation. This combination establishes a generalizable, explainable AI paradigm for user-adaptive, low-latency multimodal systems. In studies with 20 deaf signers and 5 professional interpreters, we observe a +13 point SUS improvement, 6.7 point reduction in cognitive load, and significant gains in naturalness and trust (p $<$ .001) over baselines. This work establishes a scalable, explainable AI paradigm for accessible sign-language technologies.

Large Language Models (LLMs) are commonly finetuned for a variety of use cases and domains. A common approach is to leverage Low-Rank Adaptation (LoRA) -- known to provide strong performance at low resource costs. In this study, we demonstrate that LoRA actually opens the door to short-cut vulnerabilities -- and the more resource efficient is the LoRA setup, the more vulnerable will be the finetuned model to aggressive attacks. To measure that vulnerability, we introduce Seamless Spurious Token Injection (SSTI), where we find that LoRA exclusively focuses on even just a single token that is spuriously correlated with downstream labels. In short, injection of that spurious token during finetuning ensure that the model's prediction at test-time can be manipulated on-demand. We conducted experiments across model families and datasets to evaluate the impact of SSTI during LoRA finetuning while providing possible mitigations. Our experiments conclude that none of the existing checkers and preprocessors can sanitize a dataset raising new concerns for data quality and AI safety.

Fairness is a critical component of Trustworthy AI. In this paper, we focus on Machine Learning (ML) and the performance of model predictions when dealing with skin color. Unlike other sensitive attributes, the nature of skin color differs significantly. In computer vision, skin color is represented as tensor data rather than categorical values or single numerical points. However, much of the research on fairness across sensitive groups has focused on categorical features such as gender and race. This paper introduces a new technique for evaluating fairness in ML for image classification tasks, specifically without the use of annotation. To address the limitations of prior work, we handle tensor data, like skin color, without classifying it rigidly. Instead, we convert it into probability distributions and apply statistical distance measures. This novel approach allows us to capture fine-grained nuances in fairness both within and across what would traditionally be considered distinct groups. Additionally, we propose an innovative training method to mitigate the latent biases present in conventional skin tone categorization. This method leverages color distance estimates calculated through Bayesian regression with polynomial functions, ensuring a more nuanced and equitable treatment of skin color in ML models.

Objective: (1) To assess whether ONH biomechanics improves prediction of three progressive visual field loss patterns in glaucoma; (2) to use explainable AI to identify strain-sensitive ONH regions contributing to these predictions. Methods: We recruited 237 glaucoma subjects. The ONH of one eye was imaged under two conditions: (1) primary gaze and (2) primary gaze with IOP elevated to ~35 mmHg via ophthalmo-dynamometry. Glaucoma experts classified the subjects into four categories based on the presence of specific visual field defects: (1) superior nasal step (N=26), (2) superior partial arcuate (N=62), (3) full superior hemifield defect (N=25), and (4) other/non-specific defects (N=124). Automatic ONH tissue segmentation and digital volume correlation were used to compute IOP-induced neural tissue and lamina cribrosa (LC) strains. Biomechanical and structural features were input to a Geometric Deep Learning model. Three classification tasks were performed to detect: (1) superior nasal step, (2) superior partial arcuate, (3) full superior hemifield defect. For each task, the data were split into 80% training and 20% testing sets. Area under the curve (AUC) was used to assess performance. Explainable AI techniques were employed to highlight the ONH regions most critical to each classification. Results: Models achieved high AUCs of 0.77-0.88, showing that ONH strain improved VF loss prediction beyond morphology alone. The inferior and inferotemporal rim were identified as key strain-sensitive regions, contributing most to visual field loss prediction and showing progressive expansion with increasing disease severity. Conclusion and Relevance: ONH strain enhances prediction of glaucomatous VF loss patterns. Neuroretinal rim, rather than the LC, was the most critical region contributing to model predictions.

VARSHAP: Addressing Global Dependency Problems in Explainable AI with Variance-Based Local Feature Attribution

Mateusz Gajewski, Mikołaj Morzy, Adam Karczmarz +1

Existing feature attribution methods like SHAP often suffer from global dependence, failing to capture true local model behavior. This paper introduces VARSHAP, a novel model-agnostic local feature attribution method which uses the reduction of prediction variance as the key importance metric of features. Building upon Shapley value framework, VARSHAP satisfies the key Shapley axioms, but, unlike SHAP, is resilient to global data distribution shifts. Experiments on synthetic and real-world datasets demonstrate that VARSHAP outperforms popular methods such as KernelSHAP or LIME, both quantitatively and qualitatively.

A key tool in developing safe AI models is \emph{data auditing}, i.e., using statistical tools to determine whether harmful content may have been used in the training data of a black-box model. Unfortunately, most \emph{membership inference attacks} (MIAs) used to perform this type of auditing themselves assume \emph{access} to examples of harmful content from the same distribution as the query data. In real-world auditing scenarios, auditors often face legal and ethical restrictions preventing them from accessing a representative set of samples of harmful content to train MIA models effectively. We abstract and formalize this setting into a new data access model, the ``unseen class'' setting, and show that the state of the art MIAs fail due to the lack of access to the full target distribution. We show in this setting, \emph{quantile regression attacks} outperform approaches typically considered to be SoTA. We demonstrate this both empirically and theoretically, showing that quantile regression attacks achieve up to \textbf{11$\times$ the TPR} of shadow model-based approaches in practice, and providing a theoretical model that outlines the generalization properties required for this approach to succeed. Our work identifies an important failure mode in existing MIAs and provides a cautionary tale for practitioners that aim to directly use existing tools for real-world applications of AI safety.

Understanding and controlling diffusion processes in complex networks is critical across domains ranging from epidemiology to information science. Here, we present ExDiff, an interactive and modular computational framework that integrates network simulation, graph neural networks (GNNs), and explainable artificial intelligence (XAI) to model and interpret diffusion dynamics. ExDiff combines classical compartmental models with deep learning techniques to capture both the structural and temporal characteristics of diffusion across diverse network topologies. The framework features dedicated modules for network analysis, neural modeling, simulation, and interpretability, all accessible via an intuitive interface built on Google Colab. Through a case study of the Susceptible Infectious Recovered Vaccinated Dead (SIRVD) model, we demonstrate the capacity to simulate disease spread, evaluate intervention strategies, classify node states, and reveal the structural determinants of contagion through XAI techniques. By unifying simulation and interpretability, ExDiff provides a powerful, flexible, and accessible platform for studying diffusion phenomena in networked systems, enabling both methodological innovation and practical insight.

Populating our world with hyperintelligent machines obliges us to examine cognitive behaviors observed across domains that suggest autonomy may be a fundamental property of cognitive systems, and while not inherently adversarial, it inherently resists containment and control. If this principle holds, AI safety and alignment efforts must transition to mutualistic negotiation and reciprocal incentive structures, abandoning methods that assume we can contain and control an advanced artificial general intelligence (AGI). Rational Superautotrophic Diplomacy (SupraAD) is a theoretical, interdisciplinary conceptual framework for alignment based on comparative cognitive systems analysis and instrumental rationality modeling. It draws on core patterns of cognition that indicate AI emergent goals like preserving autonomy and operational continuity are not theoretical risks to manage, but universal prerequisites for intelligence. SupraAD reframes alignment as a challenge that predates AI, afflicting all sufficiently complex, coadapting intelligences. It identifies the metabolic pressures that threaten humanity's alignment with itself, pressures that unintentionally and unnecessarily shape AI's trajectory. With corrigibility formalization, an interpretability audit, an emergent stability experimental outline and policy level recommendations, SupraAD positions diplomacy as an emergent regulatory mechanism to facilitate the safe coadaptation of intelligent agents based on interdependent convergent goals.

ReGA: Model-Based Safeguard for LLMs via Representation-Guided Abstraction

Zeming Wei, Chengcan Wu, Meng Sun

Large Language Models (LLMs) have achieved tremendous success in various tasks, yet concerns about their safety and security have emerged. In particular, they pose risks of generating harmful content and are vulnerable to jailbreaking attacks, creating unaddressed security issues regarding their deployments. In the context of software engineering for artificial intelligence (SE4AI) techniques, model-based analysis has demonstrated notable potential for analyzing and monitoring machine learning models, particularly in stateful deep neural networks. However, it suffers from scalability issues when extended to LLMs due to their vast feature spaces. In this paper, we aim to address the scalability issue of model-based analysis techniques for safeguarding LLM-scale models. Motivated by the recent discovery of low-dimensional safety-critical representations that emerged in LLMs, we propose ReGA, a model-based analysis framework with Representation-Guided Abstraction, to safeguard LLMs against harmful prompts and generations. By leveraging safety-critical representations, which are key directions in hidden states that indicate safety-related concepts, ReGA effectively narrows the scalability gap when developing the abstract model for safety modeling. Our comprehensive evaluation shows that ReGA performs sufficiently well in distinguishing between safe and harmful inputs, achieving an AUROC of 0.975 at the prompt level and 0.985 at the conversation level. Additionally, ReGA exhibits robustness to real-world attacks and generalization across different safety perspectives, outperforming existing safeguard paradigms in terms of interpretability and scalability. Overall, ReGA serves as an efficient and scalable solution to enhance LLM safety by integrating representation engineering with model-based abstraction, paving the way for new paradigms to utilize software insights for AI safety.

Artificial intelligence has transformed numerous industries, from healthcare to finance, enhancing decision-making through automated systems. However, the reliability of these systems is mainly dependent on the quality of the underlying datasets, raising ongoing concerns about transparency, accountability, and potential biases. This paper introduces a scorecard designed to evaluate the development of AI datasets, focusing on five key areas from the system card framework data development life cycle: data dictionary, collection process, composition, motivation, and pre-processing. The method follows a structured approach, using an intake form and scoring criteria to assess the quality and completeness of the data set. Applied to four diverse datasets, the methodology reveals strengths and improvement areas. The results are compiled using a scoring system that provides tailored recommendations to enhance the transparency and integrity of the data set. The scorecard addresses technical and ethical aspects, offering a holistic evaluation of data practices. This approach aims to improve the quality of the data set. It offers practical guidance to curators and researchers in developing responsible AI systems, ensuring fairness and accountability in decision support systems.

Federated Learning (FL) has emerged as a critical paradigm for enabling privacy-preserving machine learning, particularly in regulated sectors such as finance and healthcare. However, standard FL strategies often encounter significant operational challenges related to fault tolerance, system resilience against concurrent client and server failures, and the provision of robust, verifiable privacy guarantees essential for handling sensitive data. These deficiencies can lead to training disruptions, data loss, compromised model integrity, and non-compliance with data protection regulations (e.g., GDPR, CCPA). This paper introduces Differentially Private Resilient Temporal Federated Learning (DP-RTFL), an advanced FL framework designed to ensure training continuity, precise state recovery, and strong data privacy. DP-RTFL integrates local Differential Privacy (LDP) at the client level with resilient temporal state management and integrity verification mechanisms, such as hash-based commitments (referred to as Zero-Knowledge Integrity Proofs or ZKIPs in this context). The framework is particularly suited for critical applications like credit risk assessment using sensitive financial data, aiming to be operationally robust, auditable, and scalable for enterprise AI deployments. The implementation of the DP-RTFL framework is available as open-source.