arXiv subjects

Valerio Pascucci

Publications and source records attributed to Valerio Pascucci.

At least 19 recordsLinked to original sources

arXiv2026

Multi-Agent Discovery and Resource-Aware Autonomous Exploration of Scientific Datasets

Modern scientific facilities and instruments generate datasets at scales that are difficult for individual researchers to discover, access, and explore. Although many datasets are publicly available, using them often requires familiarity with repository organization, data formats, multiresolution structures, and visualization parameters. We present WebVisus, a constrained and resource-aware multi-agent system for discovering and autonomously exploring remote, multiresolution scientific datasets. Given a natural-language research question, WebVisus identifies the user's intent and launches an autonomous exploration agent that examines slices, volumes, and timesteps while adapting data resolution and retrieval quality to available client memory and computational resources. This design supports progressive exploration without complete dataset downloads or manual configuration of low-level visualization parameters using natural languages. We report the system architecture, constrained agent protocol, resource-aware access mechanism, and case studies evaluating autonomous visual exploration and resource-aware agentic access across scientific datasets.

cs.HC↗

arXiv2026

A Query-Efficient Stochastic Volume Rendering Framework for Time-Varying Implicit Neural Volumes

Time-varying implicit neural representations (INRs) provide a compact representation of scientific volumes and, for modalities such as dynamic X-ray computed tomography (CT), are often the only practical way to represent the data. However, interactive volume rendering of INRs is challenging, as cheap memory lookups are replaced by expensive neural inferences, hindering the performance. Therefore, conventional volume rendering methods such as ray marching with dense sampling are often impractical. While resampling, caching, and retraining can mitigate this cost, they compromise convenience and accuracy and become impractical for time-varying data. We tackle these challenges using a query-efficient stochastic volume rendering framework based on delta tracking. Our system employs a four-stage pipeline that exploits heterogeneous parallelism, using ray tracing cores for traversal and tensor cores for batched neural evaluation. Furthermore, we present strategies to reduce INR queries via ray budgeting and query pruning, thereby increasing per-frame performance. Using our renderer, many time-varying INRs can be rendered directly from their original representation. The system achieves ~30-40 FPS at 1024x1024 resolution on an RTX 4090 GPU and converges to high-fidelity images. Moreover, the system enables interactive temporal exploration of the continuous domain, with timestep updates taking approximately 1-2 ms.

cs.GR↗

arXiv2026

Quantifying the Impact of Lossy Compression on Neural Generative Surrogate Modeling

Neural networks are used as generative surrogate models for scientific discovery, which are trainable approximations of scientific simulations. These models enable users to replace time-consuming numerical simulations with learned alternatives, providing quick solutions. However, high-fidelity generative surrogate models require massive training datasets, which can create storage and I/O challenges. Lossy compression is a promising way to reduce this burden, but compression errors may affect the model quality in subtle ways, making it challenging to quantify their impact. In this work, we examine how lossy compression of training data impacts the quality of generative surrogate models. We begin by characterizing the uncertainty inherent in training neural networks, showing that identical training configurations can produce different models. By exploiting this variability, we propose a method to estimate how much compression-induced error a surrogate model can tolerate without affecting its accuracy. Evaluation of two application simulations demonstrates that our approach significantly reduces memory/storage requirements and speeds up training while producing high-quality surrogate models. These results show that lossy compression saves data storage up to 23.7x and 39x with negligible impact on the quality of the surrogate model. Meanwhile, reducing the size of the training data set also enhances the data loading speed and reduces the training time by up to 3x.

cs.DC↗

arXiv2026

RAFI -- A Ray/Work Forwarding Infrastructure for Data Parallel Multi-Node/Multi-GPU Computing

We present RaFI, a CUDA and MPI based software framework that simplifies the task of building GPU-enabled data-parallel software where rays or similar work items need to migrate between different GPUs. RaFI provides a simple interface for CUDA kernels to forward such work items to other GPUs, while under the hood managing all the CUDA and MPI related work required to make this happen. We describe RaFI's motivation and implementation, and show its potential in several example applications.

cs.DC↗

arXiv2026

PC-MIL: Decoupling Feature Resolution from Supervision Scale in Whole-Slide Learning

Whole-slide image (WSI) classification in computational pathology is commonly formulated as slide-level Multiple Instance Learning (MIL) with a single global bag representation. However, slide-level MIL is fundamentally underconstrained: optimizing only global labels encourages models to aggregate features without learning anatomically meaningful localization. This creates a mismatch between the scale of supervision and the scale of clinical reasoning. Clinicians assess tumor burden, focal lesions, and architectural patterns within millimeter-scale regions, whereas standard MIL is trained only to predict whether "somewhere in the slide there is cancer." As a result, the model's inductive bias effectively erases anatomical structure. We propose Progressive-Context MIL (PC-MIL), a framework that treats the spatial extent of supervision as a first-class design dimension. Rather than altering magnification, patch size, or introducing pixel-level segmentation, we decouple feature resolution from supervision scale. Using fixed 20x features, we vary MIL bag extent in millimeter units and anchor supervision at a clinically motivated 2mm scale to preserve comparable tumor burden and avoid confounding scale with lesion density. PC-MIL progressively mixes slide- and region-level supervision in controlled proportions, enabling explicit train-context x test-context analysis. On 1,476 prostate WSIs from five public datasets for binary cancer detection, we show that anatomical context is an independent axis of generalization in MIL, orthogonal to feature resolution: modest regional supervision improves cross-context performance, and balanced multi-context training stabilizes accuracy across slide and regional evaluation without sacrificing global performance. These results demonstrate that supervision extent shapes MIL inductive bias and support anatomically grounded WSI generalization.

cs.CV↗

arXiv2026

SIMPLER: H&E-Informed Representation Learning for Structured Illumination Microscopy

Structured Illumination Microscopy (SIM) enables rapid, high-contrast optical sectioning of fresh tissue without staining or physical sectioning, making it promising for intraoperative and point-of-care diagnostics. Recent foundation and large-scale self-supervised models in digital pathology have demonstrated strong performance on section-based modalities such as Hematoxylin and Eosin (H&E) and immunohistochemistry (IHC). However, these approaches are predominantly trained on thin tissue sections and do not explicitly address thick-tissue fluorescence modalities such as SIM. When transferred directly to SIM, performance is constrained by substantial modality shift, and naive fine-tuning often overfits to modality-specific appearance rather than underlying histological structure. We introduce SIMPLER (Structured Illumination Microscopy-Powered Learning for Embedding Representations), a cross-modality self-supervised pretraining framework that leverages H&E as a semantic anchor to learn reusable SIM representations. H&E encodes rich cellular and glandular structure aligned with established clinical annotations, while SIM provides rapid, nondestructive imaging of fresh tissue. During pretraining, SIM and H&E are progressively aligned through adversarial, contrastive, and reconstruction-based objectives, encouraging SIM embeddings to internalize histological structure from H&E without collapsing modality-specific characteristics. A single pretrained SIMPLER encoder transfers across multiple downstream tasks, including multiple instance learning and morphological clustering, consistently outperforming SIM models trained from scratch or H&E-only pretraining. These results suggest that histology-guided cross-modal pretraining yields biologically grounded SIM embeddings suitable for broad downstream reuse.

cs.CV↗

arXiv2026

Animating Petascale Time-varying Data on Commodity Hardware with LLM-assisted Scripting

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.

cs.AI↗

arXiv2026

Large Data Acquisition and Analytics at Synchrotron Radiation Facilities

Synchrotron facilities like the Cornell High Energy Synchrotron Source (CHESS) generate massive data volumes from complex beamline experiments, but face challenges such as limited access time, the need for on-site experiment monitoring, and managing terabytes of data per user group. We present the design, deployment, and evaluation of a framework that addresses CHESS's data acquisition and management issues. Deployed on a secure CHESS server, our system provides real time, web-based tools for remote experiment monitoring and data quality assessment, improving operational efficiency. Implemented across three beamlines (ID3A, ID3B, ID4B), the framework managed 50-100 TB of data and over 10 million files in late 2024. Testing with 43 research groups and 86 dashboards showed reduced overhead, improved accessibility, and streamlined data workflows. Our paper highlights the development, deployment, and evaluation of our framework and its transformative impact on synchrotron data acquisition.

cs.HC↗

arXiv2025

Climate Data for Power Systems Applications: Lessons in Reusing Wildfire Smoke Data for Solar PV Studies

Data reuse is using data for a purpose distinct from its original intent. As data sharing becomes more prevalent in science, enabling effective data reuse is increasingly important. In this paper, we present a power systems case study of data repurposing for enabling data reuse. We define data repurposing as the process of transforming data to fit a new research purpose. In our case study, we repurpose a geospatial wildfire smoke forecast dataset into a historical dataset. We analyze its efficacy toward analyzing wildfire smoke impact on solar photovoltaic energy production. We also provide documentation and interactive demos for using the repurposed dataset. We identify key enablers of data reuse including metadata standardization, contextual documentation, and communication between data creators and reusers. We also identify obstacles to data reuse such as risk of misinterpretation and barriers to efficient data access. Through an iterative approach to data repurposing, we demonstrate how leveraging and expanding knowledge transfer infrastructures like online documentation, interactive visualizations, and data streaming directly address these obstacles. The findings facilitate big data use from other domains for power systems applications and grid resiliency.

cs.HC↗

arXiv2025

The Making of a Community Dark Matter Dataset with the National Science Data Fabric

Dark matter is believed to constitute approximately 85 percent of the universes matter, yet its fundamental nature remains elusive. Direct detection experiments, though globally deployed, generate data that is often locked within custom formats and non-reproducible software stacks, limiting interdisciplinary analysis and innovation. This paper presents a collaboration between the National Science Data Fabric (NSDF) and dark matter researchers to improve accessibility, usability, and scientific value of a calibration dataset collected with Cryogenic Dark Matter Search (CDMS) detectors at the University of Minnesota. We describe how NSDF services were used to convert data from a proprietary format into an open, multi-resolution IDX structure; develop a web-based dashboard for easily viewing signals; and release a Python-compatible CLI to support scalable workflows and machine learning applications. These contributions enable broader use of high-value dark matter datasets, lower the barrier to entry for new collaborators, and support reproducible, cross-disciplinary research.

hep-ex↗

arXiv2025

Scalable Climate Data Analysis: Balancing Petascale Fidelity and Computational Cost

The growing resolution and volume of climate data from remote sensing and simulations pose significant storage, processing, and computational challenges. Traditional compression or subsampling methods often compromise data fidelity, limiting scientific insights. We introduce a scalable ecosystem that integrates hierarchical multiresolution data management, intelligent transmission, and ML-assisted reconstruction to balance accuracy and efficiency. Our approach reduces storage and computational costs by 99\%, lowering expenses from \$100,000 to \$24 while maintaining a Root Mean Square (RMS) error of 1.46 degrees Celsius. Our experimental results confirm that even with significant data reduction, essential features required for accurate climate analysis are preserved. Validated on petascale NASA climate datasets, this solution enables cost-effective, high-fidelity climate analysis for research and decision-making.

physics.ao-ph↗

arXiv2025

See or Recall: A Sanity Check for the Role of Vision in Solving Visualization Question Answer Tasks with Multimodal LLMs

Recent developments in multimodal large language models (MLLM) have equipped language models to reason about vision and language jointly. This permits MLLMs to both perceive and answer questions about data visualization across a variety of designs and tasks. Applying MLLMs to a broad range of visualization tasks requires us to properly evaluate their capabilities, and the most common way to conduct evaluation is through measuring a model's visualization reasoning capability, analogous to how we would evaluate human understanding of visualizations (e.g., visualization literacy). However, we found that in the context of visualization question answering (VisQA), how an MLLM perceives and reasons about visualizations can be fundamentally different from how humans approach the same problem. During the evaluation, even without visualization, the model could correctly answer a substantial portion of the visualization test questions, regardless of whether any selection options were provided. We hypothesize that the vast amount of knowledge encoded in the language model permits factual recall that supersedes the need to seek information from the visual signal. It raises concerns that the current VisQA evaluation may not fully capture the models' visualization reasoning capabilities. To address this, we propose a comprehensive sanity check framework that integrates a rule-based decision tree and a sanity check table to disentangle the effects of "seeing" (visual processing) and "recall" (reliance on prior knowledge). This validates VisQA datasets for evaluation, highlighting where models are truly "seeing", positively or negatively affected by the factual recall, or relying on inductive biases for question answering. Our study underscores the need for careful consideration in designing future visualization understanding studies when utilizing MLLMs.

cs.HC↗

arXiv2025

Approximate Puzzlepiece Compositing

The increasing demand for larger and higher fidelity simulations has made Adaptive Mesh Refinement (AMR) and unstructured mesh techniques essential to focus compute effort and memory cost on just the areas of interest in the simulation domain. The distribution of these meshes over the compute nodes is often determined by balancing compute, memory, and network costs, leading to distributions with jagged nonconvex boundaries that fit together much like puzzle pieces. It is expensive, and sometimes impossible, to re-partition the data posing a challenge for in situ and post hoc visualization as the data cannot be rendered using standard sort-last compositing techniques that require a convex and disjoint data partitioning. We present a new distributed volume rendering and compositing algorithm, Approximate Puzzlepiece Compositing, that enables fast and high-accuracy in-place rendering of AMR and unstructured meshes. Our approach builds on Moment-Based Ordered-Independent Transparency to achieve a scalable, order-independent compositing algorithm that requires little communication and does not impose requirements on the data partitioning. We evaluate the image quality and scalability of our approach on synthetic data and two large-scale unstructured meshes on HPC systems by comparing to state-of-the-art sort-last compositing techniques, highlighting our approach's minimal overhead at higher core counts. We demonstrate that Approximate Puzzlepiece Compositing provides a scalable, high-performance, and high-quality distributed rendering approach applicable to the complex data distributions encountered in large-scale CFD simulations.

cs.GR↗

arXiv2024

Sustainable Data Democratization: A Multifaceted Investment for an Equitable Future

The urgent need for data democratization in scientific research was the focal point of a panel discussion at SC23 in Denver, Colorado, from November 12 to 17, 2023. This article summarizes the outcomes of that discussion and subsequent conversations. We advocate for strategic investments in financial, human, and technological resources for sustainable data democratization. Emphasizing that data is central to scientific discovery and AI deployment, we highlight barriers such as limited access, inadequate financial incentives for cross-domain collaboration, and a shortage of workforce development initiatives. Our recommendations aim to guide decision-makers in fostering an inclusive research community, breaking down research silos, and developing a skilled workforce to advance scientific discovery.

cs.DB↗

arXiv2024

Bimodal Visualization of Industrial X-Ray and Neutron Computed Tomography Data

Advanced manufacturing creates increasingly complex objects with material compositions that are often difficult to characterize by a single modality. Our collaborating domain scientists are going beyond traditional methods by employing both X-ray and neutron computed tomography to obtain complementary representations expected to better resolve material boundaries. However, the use of two modalities creates its own challenges for visualization, requiring either complex adjustments of bimodal transfer functions or the need for multiple views. Together with experts in nondestructive evaluation, we designed a novel interactive bimodal visualization approach to create a combined view of the co-registered X-ray and neutron acquisitions of industrial objects. Using an automatic topological segmentation of the bivariate histogram of X-ray and neutron values as a starting point, the system provides a simple yet effective interface to easily create, explore, and adjust a bimodal visualization. We propose a widget with simple brushing interactions that enables the user to quickly correct the segmented histogram results. Our semiautomated system enables domain experts to intuitively explore large bimodal datasets without the need for either advanced segmentation algorithms or knowledge of visualization techniques. We demonstrate our approach using synthetic examp

cs.GR↗

arXiv2024

Web-based Visualization and Analytics of Petascale data: Equity as a Tide that Lifts All Boats

Scientists generate petabytes of data daily to help uncover environmental trends or behaviors that are hard to predict. For example, understanding climate simulations based on the long-term average of temperature, precipitation, and other environmental variables is essential to predicting and establishing root causes of future undesirable scenarios and assessing possible mitigation strategies. While supercomputer centers provide a powerful infrastructure for generating petabytes of simulation output, accessing and analyzing these datasets interactively remains challenging on multiple fronts. This paper presents an approach to managing, visualizing, and analyzing petabytes of data within a browser on equipment ranging from the top NASA supercomputer to commodity hardware like a laptop. Our novel data fabric abstraction layer allows user-friendly querying of scientific information while hiding the complexities of dealing with file systems or cloud services. We also optimize network utilization while streaming from petascale repositories through state-of-the-art progressive compression algorithms. Based on this abstraction, we provide customizable dashboards that can be accessed from any device with any internet connection, enabling interactive visual analysis of vast amounts of data to a wide range of users - from top scientists with access to leadership-class computing environments to undergraduate students of disadvantaged backgrounds from minority-serving institutions. We focus on NASA's use of petascale climate datasets as an example of particular societal impact and, therefore, a case where achieving equity in science participation is critical. We further validate our approach by deploying the dashboards and simplified training materials in the classroom at a minority-serving institution.

cs.HC↗

arXiv2024

Visualization Literacy of Multimodal Large Language Models: A Comparative Study

The recent introduction of multimodal large language models (MLLMs) combine the inherent power of large language models (LLMs) with the renewed capabilities to reason about the multimodal context. The potential usage scenarios for MLLMs significantly outpace their text-only counterparts. Many recent works in visualization have demonstrated MLLMs' capability to understand and interpret visualization results and explain the content of the visualization to users in natural language. In the machine learning community, the general vision capabilities of MLLMs have been evaluated and tested through various visual understanding benchmarks. However, the ability of MLLMs to accomplish specific visualization tasks based on visual perception has not been properly explored and evaluated, particularly, from a visualization-centric perspective. In this work, we aim to fill the gap by utilizing the concept of visualization literacy to evaluate MLLMs. We assess MLLMs' performance over two popular visualization literacy evaluation datasets (VLAT and mini-VLAT). Under the framework of visualization literacy, we develop a general setup to compare different multimodal large language models (e.g., GPT4-o, Claude 3 Opus, Gemini 1.5 Pro) as well as against existing human baselines. Our study demonstrates MLLMs' competitive performance in visualization literacy, where they outperform humans in certain tasks such as identifying correlations, clusters, and hierarchical structures.

cs.CL↗

arXiv2023

AVA: Towards Autonomous Visualization Agents through Visual Perception-Driven Decision-Making

With recent advances in multi-modal foundation models, the previously text-only large language models (LLM) have evolved to incorporate visual input, opening up unprecedented opportunities for various applications in visualization. Our work explores the utilization of the visual perception ability of multi-modal LLMs to develop Autonomous Visualization Agents (AVAs) that can interpret and accomplish user-defined visualization objectives through natural language. We propose the first framework for the design of AVAs and present several usage scenarios intended to demonstrate the general applicability of the proposed paradigm. The addition of visual perception allows AVAs to act as the virtual visualization assistant for domain experts who may lack the knowledge or expertise in fine-tuning visualization outputs. Our preliminary exploration and proof-of-concept agents suggest that this approach can be widely applicable whenever the choices of appropriate visualization parameters require the interpretation of previous visual output. Feedback from unstructured interviews with experts in AI research, medical visualization, and radiology has been incorporated, highlighting the practicality and potential of AVAs. Our study indicates that AVAs represent a general paradigm for designing intelligent visualization systems that can achieve high-level visualization goals, which pave the way for developing expert-level visualization agents in the future.

cs.HC↗