Symbolic discovery of governing equations is a long-standing goal in scientific machine learning, yet a fundamental trade-off persists between interpretability and scalable learning. Classical symbolic regression methods yield explicit analytic expressions but rely on combinatorial search, whereas neural networks scale efficiently with data and dimensionality but produce opaque representations. In this work, we introduce Symbolic Kolmogorov-Arnold Networks (Symbolic-KANs), a neural architecture that bridges this gap by embedding discrete symbolic structure directly within a trainable deep network. Symbolic-KANs represent multivariate functions as compositions of learned univariate primitives applied to learned scalar projections, guided by a library of analytic primitives, hierarchical gating, and symbolic regularization that progressively sharpens continuous mixtures into one-hot selections. After gated training and discretization, each active unit selects a single primitive and projection direction, yielding compact closed-form expressions without post-hoc symbolic fitting. Symbolic-KANs further act as scalable primitive discovery mechanisms, identifying the most relevant analytic components that can subsequently inform candidate libraries for sparse equation-learning methods. We demonstrate that Symbolic-KAN reliably recovers correct primitive terms and governing structures in data-driven regression and inverse dynamical systems. Moreover, the framework extends to forward and inverse physics-informed learning of partial differential equations, producing accurate solutions directly from governing constraints while constructing compact symbolic representations whose selected primitives reflect the true analytical structure of the underlying equations. These results position Symbolic-KAN as a step toward scalable, interpretable, and mechanistically grounded learning of governing laws.

Generative counterfactuals offer a promising avenue for explainable AI in medical imaging, yet ensuring these synthesized images are both anatomically faithful and clinically effective remains a significant challenge. This work presents a domain-specific diffusion framework for generating ”healthy” counterfactuals from chest X-rays with cardiomegaly, underpinned by a systematic metric-guided inference strategy. In contrast to methods relying on static sampling parameters, our approach iteratively explores the inference hyperparameter space to maximize our composite selection criterion, CF Score, that integrates our novel Faithfulness-Effectiveness Trade-off (F ET ) metric.

We extend the evaluation of counterfactual utility beyond simple classification shifts by conducting the simultaneous validation against radiologist annotations and eye-tracking data. Using the REFLACX dataset, we demonstrate that difference maps derived from our counterfactuals exhibit strong spatial alignment with expert visual attention and annotation. Quantified by Normalized Cross-Correlation, Hit Rate, pixel-wise ROC-AUC, and AUC-IoU, our results confirm that metric-guided counterfactuals provide dense and clinically relevant localizations of pathology that closely mirror human diagnostic reasoning.

High-Dimensional Low-Sample Size (HDLSS) tabular domains (e.g., omics) are characterized by n ≪ m, where n = number of samples, and m = number of features. Such domains often exhibit strong local correlation groups, sparse cross-group dependencies, heavy-tailed non-Gaussian marginals, heteroscedastic noise, and structured missingness, making direct density learning in Rm ill-conditioned since n ≪ m. We propose BSTabDiff, a block-subunit generative framework that partitions the m observed features into M latent blocks (M ≪ m) and generates each block via a shared low-dimensional subunit variable, concentrating global dependence learning in the compact block-latent space RM while decoding to the full feature space with copula-driven dependence, flexible per-feature marginals, and explicit missingness mechanisms. BSTabDiff supports modern deep priors on block latents, including diffusion and normalizing flows, enabling stable synthesis and controllable benchmark generation in the HDLSS regime. Empirically, BSTabDiff produces more realistic and stable high-dimensional synthetic data when compared with unstructured tabular generators on HDLSS data.

High-dimensional tabular data lacks a natural feature order, limiting the applicability of permutation-sensitive deep learning models. We propose DynaTab, a dynamic feature ordering-enabled architecture inspired by neural rewiring. We introduce a lightweight criterion that predicts when feature permutation will benefit a dataset by quantifying its intrinsic complexity. DynaTab dynamically reorders features via a neural rewiring algorithm and processes them through a compact, dynamic order-aware combination of separate learned positional embedding, importance-based gating, and masked attention layers, compatible with any sequence-sensitive backbone. Trained end-to-end with bespoke dynamic feature ordering (DFO) and dispersion losses, DynaTab achieves statistically significant gains, particularly on high-dimensional datasets, where it is benchmarked against 45 state-of-the-art baselines across 36 different real-world tabular datasets. Our results position DynaTab as a compelling new paradigm for high-dimensional tabular deep learning.

Interpersonal communication in data science can yield sought-after insights, but presentation environments are often not conducive for live analysis, forcing the process to move offline. Through a formative survey with 16 participants, we identified both technical (e.g., complexity of tools) and psychological (e.g., pressure of programming during presentation) factors constraining live data analysis. To enable live analysis, we present SlideSAVR, a data-driven presentation assistant that leverages sketching and voice inputs in live discussion to support collaborative data analysis during presentations. Powered by an agentic framework that flexibly defines augmentation rules, updates slide content dynamically to match the live context, and automates backend computations, SlideSAVR enables fluid audience-presenter interaction and reduces the need for offline reanalysis and follow-up communication. We demonstrate SlideSAVR’s ability to support a range of tasks through nine representative use cases. We further evaluate the system’s accuracy and computation time across different settings, showing that SlideSAVR can reliably perform diverse tasks when provided with both sketch and voice inputs.

Time series forecasting is a fundamental tool with wide ranging applications, yet recent debates question whether complex nonlinear architectures truly outperform simple linear models. Prior claims of dominance of the linear model often stem from benchmarks that lack diverse temporal dynamics and employ biased evaluation protocols. We revisit this debate through TimeSynth, a structured framework that emulates key properties of real world time series,including non-stationarity, periodicity, trends, and phase modulation by creating synthesized signals whose parameters are derived from real-world time series. Evaluating four model families Linear, Multi Layer Perceptrons (MLP), Convolutional Neural Networks (CNNs), and Transformers, we find a systematic bias in linear models: they collapse to simple oscillation regardless of signal complexity. Nonlinear models avoid this collapse and gain clear advantages as signal complexity increases. Notably, Transformers and CNN based models exhibit slightly greater adaptability to complex modulated signals compared to MLPs. Beyond clean forecasting, the framework highlights robustness differences under distribution and noise shifts and removes biases of prior benchmarks by using independent instances for train, test, and validation for each signal family. Collectively, TimeSynth provides a principled foundation for understanding when different forecasting approaches succeed or fail, moving beyond oversimplified claims of model equivalence. 

While medical Vision-Language models (VLMs) achieve strong performance on tasks such as tumor or organ segmentation and diagnosis prediction, their opaque latent representations limit clinical trust and the ability to explain predictions. Interpretability of these multimodal representations are therefore essential for the trustworthy clinical deployment of pretrained medical VLMs. However, current interpretability methods, such as gradient- or attention-based visualizations, are often limited to specific tasks such as classification. Moreover, they do not provide concept-level explanations derived from shared pretrained representations that can be reused across downstream tasks. We introduce MedConcept, a framework that uncovers latent medical concepts in a fully unsupervised manner and grounds them in clinically verifiable textual semantics. MedConcept identifies sparse neuron-level concept activations from pretrained VLM representations and translates them into pseudo-report-style summaries, enabling physician-level inspection of internal model reasoning. To address the lack of quantitative evaluation in concept-based interpretability, we introduce a quantitative semantic verification protocol that leverages an independent pretrained medical LLM as a frozen external evaluator to assess concept alignment with radiology reports. We define three concept scores, Aligned, Unaligned, and Uncertain, to quantify semantic support, contradiction, or ambiguity relative to radiology reports and use them exclusively for post hoc evaluation. These scores provide a quantitative baseline for assessing interpretability in medical VLMs.

Hyper-differential sensitivity analysis with respect to model discrepancy was recently developed to enable uncertainty quantification for optimization problems. The approach consists of two primary steps: (i) Bayesian calibration of the discrepancy between high- and low-fidelity models, and (ii) propagating the model discrepancy uncertainty through the optimization problem. When high-fidelity model evaluations are limited, as is common in practice, the prior discrepancy distribution plays a crucial role in the uncertainty analysis. However, specification of this prior is challenging due to its mathematical complexity and many hyper-parameters. This article presents a novel approach to specify the prior distribution. Our approach consists of two parts: (1) an algorithmic initialization of the prior hyper-parameters that uses existing data to initialize a hyper-parameter estimate, and (2) a visualization framework to systematically explore properties of the prior and guide tuning of the hyper-parameters to ensure that the prior captures the appropriate range of uncertainty. We provide detailed mathematical analysis and a collection of numerical examples that elucidate properties of the prior that are crucial to ensure uncertainty quantification.

Quantum processors are being integrated into HPC ecosystems as co-processors, where compilation of quantum circuits into hardware-executable form determines both output fidelity and runtime. Current compilers use a fixed pass sequence and ignore the fact that optimal pass selection varies with circuit, hardware, and noise conditions. We present TuniQ, a reinforcement learning-based system that selects compilation passes at each pipeline stage, adapting to circuit, backend, and current noise profile. TuniQ introduces several novel design components like a dual-encoder for stage-aware representation, shaped rewards for cross-stage credit assignment, and dynamic action masking for valid compilation. Evaluated across diverse quantum workloads on multiple IBM Quantum Cloud processors, TuniQ improves fidelity and reduces compilation time over the state-of-the-art IBM Qiskit transpiler, generalizes across backends without retraining, and scales strongly to utility-scale circuits with growing advantage.

Accelerated 3D late gadolinium enhancement (LGE) MRI requires robust reconstruction methods to recover thin atrial structures from undersampled k-space data. While unrolled model-based networks effectively integrate physics-driven data consistency with learned priors, they operate at the acquired resolution and may fail to fully recover high-frequency detail. We propose a hybrid unrolled reconstruction framework in which an Enhanced Deep Super-Resolution (EDSR) network replaces the proximal operator within each iteration of the optimization loop, enabling joint super-resolution enhancement and data consistency enforcement. The model is trained end-to-end on retrospectively undersampled preclinical 3D LGE datasets and compared against compressed sensing, Model-Based Deep Learning (MoDL), and self-guided Deep Image Prior (DIP) baselines. Across acceleration factors, the proposed method consistently improves PSNR and SSIM over standard unrolled reconstruction and better preserves fine cardiac structures, leading to improved LA (left atrium) segmentation performance. These results demonstrate that integrating super-resolution priors directly within model-based reconstruction provides measurable gains in accelerated 3D LGE MRI.

Mean flow (MeanFlow) enables efficient, high-fidelity image generation, yet its single-function evaluation (1-NFE) generation often cannot yield compelling results. We address this issue by introducing RMFlow, an efficient multimodal generative model that integrates a coarse 1-NFE MeanFlow transport with a subsequent tailored noise-injection refinement step. RMFlow approximates the average velocity of the flow path using a neural network trained with a new loss function that balances minimizing the Wasserstein distance between probability paths and maximizing sample likelihood. RMFlow achieves near state-of-the-art results on text-to-image, context-to-molecule, and time-series generation using only 1-NFE, at a computational cost comparable to the baseline MeanFlows. 

Patient-specific finite element models of hip cartilage and labrum (chondrolabral) mechanics have improved the understanding of form-function relationships underpinning hip osteoarthritis. While these models often assume the pelvis and femur are rigid bodies to reduce development and computational time, contact stresses may be overestimated when the pelvis and femur cannot deform. Recent advancements in element formulations, constitutive models, and integration of patient-specific boundary/loading conditions warrant a re-examination of this rigid body assumption. We assessed the influence of material representations for the pelvis and femur on finite element predictions of chondrolabral mechanics during level walking and squatting. Four material representations were evaluated: (1) rigid bodies, (2) deformable-inhomogeneous with CT-derived material properties, (3) deformable-two-material with distinct trabecular and cortical bone materials, and (4) deformable-one-material model. Patient-specific kinematics and joint reaction forces for level walking and squatting were derived from motion capture and musculoskeletal models, respectively. During level walking, there were no significant differences between bone material representations for maximum cartilage contact pressure, contact area, shear stress, or first principal strain. During squatting, rigid and one-material models produced significantly higher maximum contact pressures, while other metrics remained unaffected. Rigid models required significantly less computational runtime and memory than deformable models. Our findings indicate rigid bone assumptions suffice for level walking, improving efficiency in large cohort studies, but deformable bone models are likely warranted for activities that produce larger bone deformations such as squatting. This study provides guidance for selecting bone material representations that balance accuracy and computational efficiency in hip biomechanical analyses.

Scientists face significant visualization challenges as time-varying datasets grow in speed and volume, often requiring specialized infrastructure and expertise to handle massive datasets. Petascale climate models generated in NASA laboratories require a dedicated group of graphics and media experts and access to high-performance computing resources. Scientists may need to share scientific results with the community iteratively and quickly. However, the time-consuming trial-and-error process incurs significant data transfer overhead and far exceeds the time and resources allocated for typical post-analysis visualization tasks, disrupting the production workflow. Our paper introduces a user-friendly framework for creating 3D animations of petascale, time-varying data on a commodity workstation. Our contributions: (i) Generalized Animation Descriptor (GAD) with a keyframe-based adaptable abstraction for animation, (ii) efficient data access from cloud-hosted repositories to reduce data management overhead, (iii) tailored rendering system, and (iv) an LLM-assisted conversational interface as a scripting module to allow domain scientists with no visualization expertise to create animations of their region of interest. We demonstrate the framework's effectiveness with two case studies: first, by generating animations in which sampling criteria are specified based on prior knowledge, and second, by generating AI-assisted animations in which sampling parameters are derived from natural-language user prompts. In all cases, we use large-scale NASA climate-oceanographic datasets that exceed 1PB in size yet achieve a fast turnaround time of 1 minute to 2 hours. Users can generate a rough draft of the animation within minutes, then seamlessly incorporate as much high-resolution data as needed for the final version.

Hydrothermal plumes are turbulent structures of intense heat and mineral smoke that rise and disperse into the deep ocean. Existing models generally characterize these systems as a single axisymmetric plume originating from a point source. However, this assumption breaks down in weakly venting, spatially distributed systems, where low-flux discharge and strong background currents produce faint and distorted plumes that traditional centerline-based diagnostics fail to characterize. We present a topology-based visual analysis framework that treats plumes as time-varying three-dimensional scalar fields and captures their full structural variability. Using merge trees, we cluster plume snapshots into representative morphologies and uncover a clear coupling between plume structure and phases of the tidal cycle. To support exploration, we introduce a radial merge tree visualization that arranges plume similarity on a tidal clock, revealing periodic behaviors not discernible with existing techniques. We further develop an extremum graph based workflow that identifies discrete surface venting regions within a reconstructed porous rock, and use streamline visualizations to link subsurface transport pathways to the overlying plume.

Statistical shape modeling (SSM) is central to population level analysis of anatomical variability, yet most existing approaches rely on densely annotated segmentations and fixed latent representations. These requirements limit scalability and reduce flexibility when modeling complex anatomical variation. We introduce MorphoFlow, a sparse supervised generative shape modeling framework that learns compact probabilistic shape representations directly from sparse surface annotations. MorphoFlow integrates neural implicit shape representations with an autodecoder formulation and autoregressive normalizing flows to learn an expressive probabilistic density over the latent shape space. The neural implicit representation enables resolution-agnostic modeling of 3D anatomy, while the autodecoder formulation supports direct optimization of per-instance latent codes under sparse supervision. The autoregressive flow captures the distribution of latent anatomical variability providing a tractable, likelihood-based generative model of shapes. To promote compact and structured latent representations, we incorporate adaptive latent relevance weighting through sparsity-inducing priors, enabling the model to regulate the contribution of individual latent dimensions according to their relevance to the underlying anatomical variation while preserving generative expressivity. The resulting latent space supports uncertainty quantification and anatomically plausible shape synthesis without manual latent dimensionality tuning. Evaluation on publicly available lumbar vertebrae and femur datasets demonstrates accurate high-resolution reconstruction from sparse inputs and recovery of structured modes of anatomical variation consistent with population level trends.

The conformational activation of α_IIbβ₃integrin is crucial for platelet aggregation, a central event in hemostasis and thrombosis. Although activation can be triggered by extracellular arginine-glycine-aspartic acid (RGD)-containing ligands as well as mechanical forces, how these biochemical and mechanical cues exactly govern the structural dynamics of α_IIbβ₃remains unclear. Here, using all-atom molecular dynamics simulations, we show that mechanical force and RGD binding promote activation α_IIbβ₃through distinct mechanisms. Mechanical force applied to the RGD-binding site induces long-range, correlated motions of distant parts of the receptor, facilitating head–leg separation. In contrast, RGD binding increases localized, non-correlated fluctuations that weaken leg coordination but do not generate long-range motions. Despite these differences, both cues stabilize the open, extended conformation of α_IIbβ₃. Together, these findings suggest that mechanical and biochemical stimuli play complementary yet distinct roles in integrin conformational activation. A balance between global coordination and local fluctuations likely governs integrin activation in complex environments where the dominance of mechanical or biochemical cues could lead to distinct activation pathways and functional outcomes.

This study validates a finite deformation, nonlinear viscoelastic constitutive model for the collagen matrix of immature bovine articular cartilage, using reactive viscoelasticity. Tissue samples underwent proteoglycan (PG) digestion, losing more than 98% of their initial PG content to increase their hydraulic permeability. To verify that PG-digestion eliminated flow-dependent viscoelasticity, samples were subjected to a gravitational permeation experiment, demonstrating that their hydraulic permeability, k=268±152 mm⁴/N⋅s (n=8), was five orders of magnitude greater than reported for untreated cartilage. Digested cartilage plugs were subjected to unconfined compression stress relaxation (four consecutive 10% strain ramp-hold profiles) to fit the load response and extract material properties (RMSE_fit=1.86±0.61 kPa, n=8). Successful curve-fitting served as a necessary condition for validating the model. Then, a separate unconfined compression stress-relaxation test was performed on the same samples, to 40% compressive strain at the same ramp rate. The model was able to faithfully predict this experimental response using fitted material properties (RMSE_pred=3.95±1.33 kPa, with 0=stresses=155±37 kPa), providing a sufficient condition for validation in unconfined compression stress-relaxation. A computational model showed that flow-independent viscoelasticity of cartilage collagen can enhance the stress response by ~15% at fast strain rates, over flow-dependent effects. However, we estimate from prior studies that flow-independent viscoelasticity may enhance the stress response of cartilage by up to 200%, implying that PGs probably contribute significantly to the tissue?s flow-independent viscoelasticity.

Generative AI has demonstrated significant potential in creative design, enabling the rapid generation of visual content and imaginative concepts. Although deep AI models achieve effective featurization in the latent space, navigating the space remains a challenge. Current techniques, such as GANSlider and SliderSpace, use multiple sliders to generate high-dimensional vectors in generative AI's latent space. Despite applying (global) PCA to reduce the number of sliders, these approaches struggle with scalability and usability as the number of control dimensions increases. In this paper, we introduce LatentGandr, a visual analytics technique that facilitates latent space exploration by extracting locally linear dimensions from embeddings in high-dimensional latent spaces. By analyzing the topology and local curvature of the embeddings, LatentGandr automatically identifies local neighborhoods and computes their principal components using localized PCA. These local principal components are visualized as interactive image grids, allowing users to efficiently explore and control the generative process, providing an intuitive means to refine the generation of novel content and concepts. To evaluate the effectiveness of LatentGandr, we conducted a study comparing it to GANSlider, the current state-of-the-art visualization interface for generative AI models. The results offer insights into how localized exploration techniques can enhance user interaction with these models.

Finite element (FE) simulations emulating transcatheter pulmonary valve (TPV) system deployment in patient-specific right ventricular outflow tracts (RVOT) assume material properties for the RVOT and adjacent tissues. Sensitivity of the deployment to variation in RVOT material properties is unknown. Moreover, the effect of a transannular patch stiffness and location on simulated TPV deployment has not been explored. A sensitivity analysis on the material properties of a patient-specific RVOT during TPV deployment, modeled as an uncoupled HGO material, was conducted using FEBioUncertainSCI. Further, the effects of a transannular patch during TPV deployment were analyzed by considering two patch locations and four patch stiffnesses. Visualization of results and quantification were performed using custom metrics implemented in SlicerHeart and FEBio. Sensitivity analysis revealed that the shear modulus of the ground matrix (c), fiber modulus (k1), and fiber mean orientation angle (gamma) had the greatest effect on 95th %ile stress, whereas only c had the greatest effect on 95th %ile Lagrangian strain. First-order sensitivity indices contributed the greatest to the total-order sensitivity indices. Simulations using a transannular patch revealed that peak stress and strain were dependent on patch location. As stiffness of the patch increased, greater stress was observed at the interface connecting the patch to the RVOT, and stress in the patch itself increased while strain decreased. The total enclosed volume by the TPV device remained unchanged across all simulated patch cases. This study highlights that while uncertainties in tissue material properties and patch locations may influence functional outcomes, FE simulations provide a reliable framework for evaluating these outcomes in TPVR.