Search NASA⌕ Search

SEARCH · Search NASA

Results for “vision transformers”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 55 records · Page 3
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Discovering Strong Gravitational Lenses in the Dark Energy Survey with Interactive Machine Learning and Crowd-sourced Inspection with Space Warps

We conduct a search for strong gravitational lenses in the Dark Energy Survey (DES) Year 6 imaging data. We implement a pre-trained Vision Transformer (ViT) for our machine learning (ML) architecture and adopt interactive machine learning to construct a training sample with multiple classes to address common types of false positives. Our ML model reduces ∼236 million DES cutout images to 22,564 targets of interest, including ∼85% of previously reported galaxy–galaxy lens candidates discovered in DES. These targets were visually inspected by citizen scientists, who ruled out ∼90% as false positives. Of the remaining 2618 candidates, 149 were expert-classified as “definite” lenses and 516 as “probable” lenses, for a total of 665 systems, with 147 of these candidates being newly identified. Additionally, we trained a second ViT to find double-source plane lens systems, finding at least one double-source system. Our main ViT excels at identifying galaxy–galaxy lenses, consistently assigning high scores to candidates with high expert assessments. The top 800 ViT-scored images include ∼100 of our “definite” lens candidates. This selection is an order of magnitude higher in purity than previous convolutional neural-network-based lens searches and demonstrates the feasibility of applying our methodology for discovering large samples of lenses in future surveys.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2025

Exploring Black-box Adversarial Attacks on Low-rank Constrained Neural Networks

Low-rank compression has been shown as an effective tool to reduce parameter counts of convolutional and vision transformer architectures; however, low-rank training often reduces model robustness to adversarial perturbations. In this work, we explore the effects of low-rank training on black-box attacks, where attacked images are generated without knowledge of the low-rank parameters. We find that low-rank training is not sufficient as a black-box defense and can sometimes produce worse than expected as compared to baseline models. Influencing the spectrum of the low-rank models during training, which is known to increase model robustness against white-box attacks, improves black-box performance as well.

Schnake, Stefan [ORNL] (ORCID:0000000215183538)↗
DOE OSTI2025

Fourier-based three-dimensional multistage transformer for aberration correction in multicellular specimens

High-resolution tissue imaging is often compromised by sample-induced optical aberrations that degrade resolution and contrast. Although wavefront sensor-based adaptive optics (AO) can measure these aberrations, such hardware solutions are typically complex, expensive to implement and slow when serially mapping spatially varying aberrations across large fields of view. Here we introduce AOViFT (adaptive optical vision Fourier transformer)—a machine learning-based aberration sensing framework built around a three-dimensional multistage vision transformer that operates on Fourier domain embeddings. AOViFT infers aberrations and restores diffraction-limited performance in puncta-labeled specimens with substantially reduced computational cost, training time and memory footprint compared to conventional architectures or real-space networks. We validated AOViFT on live gene-edited zebrafish embryos, demonstrating its ability to correct spatially varying aberrations using either a deformable mirror or postacquisition deconvolution. By eliminating the need for the guide star and wavefront sensing hardware and simplifying the experimental workflow, AOViFT lowers technical barriers for high-resolution volumetric microscopy across diverse biological samples.

Alshaabi, Thayer [Howard Hughes Medical Institute,↗
DOE OSTI2025

DeepAndes: A Self-Supervised Vision Foundation Model for Multispectral Remote Sensing Imagery of the Andes

By mapping sites at large scales usingremotely sensed data, archaeologists can generate unique insights into long-term demographic trends, interregional social networks, and human adaptations in the past. Remote sensing surveys complement field-based approaches, and their reach can be especially great when combined with deep learning and computer vision techniques. However, conventional supervised deep learning methods face challenges in annotating fine-grained archaeological features at scale. In addition, while recent vision foundation models have shown remarkable success in learning large-scale remote sensing data with minimal annotations, most off-the-shelf solutions are designed for RGB images rather than multispectral satellite imagery, such as the eight-band data used in our study. In this article, we introduce DeepAndes, a transformer-based vision foundation model trained on three million multispectral satellite images, specifically tailored for Andean archaeology. DeepAndes incorporates a customized DINOv2 self-supervised learning algorithm optimized for eight-band multispectral imagery, marking the first foundation model designed explicitly for the Andes region. We evaluate its image understanding performance through imbalanced image classification, image instance retrieval, and pixel-level semantic segmentation tasks. Our experiments show that DeepAndes achieves superior F1 scores, mean average precision, and Dice scores in few-shot learning scenarios, significantly outperforming models trained from scratch or pretrained on smaller datasets. This underscores the effectiveness of large-scale self-supervised pretraining in archaeological remote sensing.

Guo, Junlin [Vanderbilt Univ., Nashville, TN (Unit↗
DOE OSTI2026

Pre-training Vision Models for the Classification of Alerts from Wide-field Time-domain Surveys

Modern wide-field time-domain surveys facilitate the study of transient, variable and moving phenomena by conducting image differencing and relaying alerts to their communities. Machine learning tools have been used on data from these surveys and their precursors for more than a decade, and convolutional neural networks (CNNs), which make predictions directly from input images, saw particularly broad adoption through the 2010s. Since then, continually rapid advances in computer vision have transformed the standard practices around using such models. It is now commonplace to use standardized architectures pre-trained on large corpora of everyday images (e.g., ImageNet). In contrast, time-domain astronomy studies still typically design custom CNN architectures and train them from scratch. Here, we explore the effects of adopting various pre-training regimens and standardized model architectures on the performance of alert classification. We find that the resulting models match or outperform a custom, specialized CNN like what is typically used for filtering alerts. Moreover, our results show that pre-training on galaxy images from Galaxy Zoo tends to yield better performance than pre-training on ImageNet or training from scratch. We observe that the design of standardized architectures are much better optimized than the custom CNN baseline, requiring significantly less time and memory for inference despite having more trainable parameters. On the eve of the Legacy Survey of Space and Time and other image-differencing surveys, these findings advocate for a paradigm shift in the creation of vision models for alerts, demonstrating that greater performance and efficiency, in time and in data, can be achieved by adopting the latest practices from the computer vision field.

79 ASTRONOMY AND ASTROPHYSICS↗
DOE OSTI2026

Autoregressive long-horizon prediction of plasma edge dynamics *

Accurate modeling of scrape-off layer (SOL) and divertor-edge dynamics is vital for designing plasma-facing components in fusion devices. High-fidelity edge fluid/neutral codes such as SOLPS-ITER capture SOL physics with high accuracy, but their computational cost limits broad parameter scans and long transient studies. We present transformer-based, autoregressive surrogates for efficient prediction of 2D, time-dependent plasma edge state fields. Trained on SOLPS-ITER spatiotemporal data for the KSTAR tokamak, the surrogates forecast electron temperature, electron density, and radiated power over extended horizons. We evaluate model variants trained with increasing autoregressive horizons (1–100 steps) on short- and long-horizon prediction tasks. Longer-horizon training systematically improves rollout stability and mitigates error accumulation, enabling stable predictions over hundreds to thousands of steps and reproducing key dynamical features such as the motion of high-radiation regions. Measured end-to-end wall-clock times show the surrogate is orders of magnitude faster than SOLPS-ITER, enabling rapid parameter exploration. Prediction accuracy degrades when the surrogate enters physical regimes not represented in the training dataset, motivating future work on data enrichment and physics-informed constraints. Overall, this approach provides a fast, accurate surrogate for computationally intensive plasma edge simulations, supporting rapid scenario exploration, control-oriented studies, and progress toward real-time applications in fusion devices.

autoregressive deep learning↗
DOE OSTI2024

A Scalable Real-Time Data Assimilation Framework for Predicting Turbulent Atmosphere Dynamics

AI-based foundation models like FourCastNet, GraphCast are revolutionizing weather and climate predictions but are not yet ready for operational use. Their limitation lies in the absence of a data assimilation system to incorporate real-time Earth system observations, crucial for accurately forecasting events like tropical cyclones. To overcome these obstacles, we introduce a generic real-time data assimilation framework and demonstrate its end-to-end performance on the Frontier supercomputer. This framework comprises two primary modules: an ensemble score filter (EnSF), which significantly outperforms the state-of-the-art data assimilation method, and a vision transformer-based surrogate capable of real-time adaptation through the integration of observational data. We demonstrate both the strong and weak scaling of our framework up to 1024 GPUs on the Exascale supercomputer, Frontier. Our results not only illustrate the framework's exceptional scalability on high-performance computing systems, but also demonstrate the importance of supercomputers in real-time data assimilation for weather and climate predictions.

Lu, Dan↗
DOE OSTI2025

2024 Annual Report Laboratory Directed Research & Development

This is the FY-24 LDRD annual report. The 62 projects concluded in fiscal year 2024, highlighted in this report, represent a glimpse into the extraordinary breadth and depth of leading-edge science, technology, and engineering endeavors ongoing at INL. I invite you to explore this report thoroughly, discovering first-hand how INL's LDRD portfolio not only fosters innovation but also nurtures researcher talent, propelling us closer to realizing our vision of transforming the world's energy future and safeguarding our critical infrastructure.

99 - GENERAL AND MISCELLANEOUS↗
DOE OSTI2026

FY-25 LDRD Annual Report

This is the FY-25 LDRD annual report. The 63 projects concluded in fiscal year 2025, highlighted in this report, represent a glimpse into the extraordinary breadth and depth of leading-edge science, technology, and engineering endeavors ongoing at INL. I invite you to explore this report thoroughly, discovering first-hand how INL's LDRD portfolio not only fosters innovation but also nurtures researcher talent, propelling us closer to realizing our vision of transforming the world's energy future and safeguarding our critical infrastructure.

99 - GENERAL AND MISCELLANEOUS↗
DOE OSTI2025

Contaminant Investigation and Pre‐Processing Opportunities for Textile‐To‐Textile Recycling

Millions of metric tons of textiles are landfilled or incinerated each year in the United States, with less than 1% of textiles recycled into new clothing or fabrics. To counter this trend, a growing number of companies and researchers are exploring how a circular economy can be applied to support textile‐to‐textile recycling. A significant barrier they face comes down to quickly and efficiently extracting pure feedstock material from post‐consumer garments that feature a mix of natural and synthetic fibers. Textile recyclers prefer pure feedstocks, as working with mixed sources typically means lower throughput, higher risk of equipment failure, and diminished business margins. To facilitate a circular economy for textiles, methods, and technologies are needed that can efficiently separate out materials and contaminants from end‐of‐life textiles to increase the flow of pure feedstocks to recyclers. This paper summarizes findings from interviews with a cross section of textile recyclers and from a review of literature to define basic feedstock requirements. In addition to our qualitative research, we deconstruct a bale of post‐consumer textiles and analyze them using computer‐vision imaging, Fourier transform infrared spectroscopy (FTIR), and machine learning. The resulting data are used to set system‐level design inputs for an automated contaminant removal system to process post‐consumer clothing into appropriate feedstocks for recycling. To set the system's levels for automated real‐time near‐infrared analysis, we identify the minimum percentage of primary material that any single garment in a load of used clothing must contain for the average of the full output stream to meet the target purity levels of recyclers. Here, the envisioned automated system can also address undesirable trace materials that might contaminate the processed stream by using imaging cameras coupled with artificial intelligence to identify sections of clothing for de‐trimming. Proof‐of‐concept machine learning algorithms are evaluated to locate and identify trims or garment areas with hidden contaminant materials. Integrating these methods into automated textile cutting systems can provide a cost‐effective means for increasing feedstock purity from used clothing, which can advance circularity for textiles by helping recyclers to reach production volumes and quality targets that were not possible solely with manual dismantling operations.

Parsons, Ryan [Rochester Institute of Technology, ↗
DOE OSTI2025

Distributed Cross-Channel Hierarchical Aggregation for Foundation Models

Vision-based scientific foundation models hold significant promise for advancing scientific discovery and innovation. This potential stems from their ability to aggregate images from diverse sources—such as varying physical groundings or data acquisition systems—and to learn spatio-temporal correlations using transformer architectures. However, tokenizing and aggregating images can be compute-intensive, a challenge not fully addressed by current distributed methods. In this work, we introduce the Distributed Cross-Channel Hierarchical Aggregation (D-CHAG) approach designed for datasets with a large number of channels across image modalities. Our method is compatible with any model-parallel strategy and any type of vision transformer architecture, significantly improving computational efficiency. We evaluated D-CHAG on hyperspectral imaging and weather forecasting tasks. When integrated with tensor parallelism and model sharding, our approach achieved up to a 75% reduction in memory usage and more than doubled sustained throughput on up to 1,024 AMD GPUs on the Frontier Supercomputer.

Tsaris, Aristeidis (aris) [ORNL] (ORCID:0000000277↗
DOE OSTI2024

Dynamic STEM-EELS for single-atom and defect measurement during electron beam transformations

This study introduces the integration of dynamic computer vision–enabled imaging with electron energy loss spectroscopy (EELS) in scanning transmission electron microscopy (STEM). This approach involves real-time discovery and analysis of atomic structures as they form, allowing us to observe the evolution of material properties at the atomic level, capturing transient states traditional techniques often miss. Rapid object detection and action system enhances the efficiency and accuracy of STEM-EELS by autonomously identifying and targeting only areas of interest. This machine learning (ML)–based approach differs from classical ML in that it must be executed on the fly, not using static data. We apply this technology to V-doped MoS 2 , uncovering insights into defect formation and evolution under electron beam exposure. This approach opens uncharted avenues for exploring and characterizing materials in dynamic states, offering a pathway to increase our understanding of dynamic phenomena in materials under thermal, chemical, and beam stimuli.

47 OTHER INSTRUMENTATION↗
DOE OSTI2025

Catalyst-Vision (PEM Catalyst Layer Image Analysis Tool) [SWR-25-100]

Catalyst-Vision (PEM Catalyst Layer Image Analysis Tool) provides an advanced Python-based tool, primarily designed for use in a Jupyter/Colab notebook, for the quantitative morphological analysis of pre-segmented shapes. While developed for analyzing PEM catalyst layers from microscopy, its methodology is suitable for characterizing any grayscale object provided on a uniform white background. The tool uses a robust computer vision pipeline based on the Euclidean Distance Transform and skeletonization to accurately measure local thickness and tortuosity, providing a comprehensive characterization of an object's geometry and internal texture. If you find this code useful, please cite our preprint as: Chan, Ai-Lin and Hayden, Steven and Harvey, Steven P. and Smeaton, Michelle and Okrucky, Caleb and Watt, John and Ulična, Soňa and Spurgeon, Steven and Jungjohann, Katherine and Alia, Shaun, Mechanism-informed breakdown: understanding degradation by controlling voltage hold patterns in PEM water electrolyzers. Preprint (2025).

Spurgeon, Steven [National Laboratory of the Rocki↗
DOE OSTI2025

United States Muon Collider Community White Paper for the European Strategy for Particle Physics Update

This document is being submitted to the 2024-2026 European Strategy for Particle Physics Update (ESPPU) process on behalf of the US Muon Collider community, with its preparation coordinated by the interim US Muon Collider Coordination Group. The US Muon Collider Community comprises a few hundred American scientists. The purpose of the document is to inform ESPPU about the US plans for Muon Collider research and development (R&D), explain how these efforts align with the broader international R&D initiatives, and present the US community vision for the future realization of this transformative project.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2026

Building an AI-native Research Ecosystem for Experimental Particle Physics: A Community Vision

Experimental particle physics seeks to understand the universe by probing its fundamental particles and forces and exploring how they govern the large-scale processes that shape cosmic evolution. This whitepaper presents a vision for how Artificial Intelligence (AI) can accelerate discovery in this field. We outline grand challenges that must be addressed to enable transformative breakthroughs and describe how current and planned experimental facilities can implement this vision to advance our understanding of the vast and complex physical world from the smallest to the largest scales. We show how facilities currently under construction, such as the HL-LHC, DUNE and soon EIC, can both benefit from and serve as proving grounds for this vision, while also enabling a longer-term goal for how future experiments -- like FCC-ee at CERN, IceCube-Gen2, a Muon Collider in the U.S., and smaller to mid-scale projects -- can be fully AI-native. We describe how a truly national-scale collaboration, jointly managed across large funding partners, and involving both DOE laboratories and universities, can make this happen.

Aarrestad, Thea Klaeboe [ETH, Zurich (main)]↗
DOE OSTI2025

Barrierless nucleation in glassy precursors drives zeolite formation

Zeolites are crystalline, microporous silicates widely used in catalysis and separations, yet the molecular mechanisms of their formation remain unresolved. Experiments indicate that hydrothermal synthesis of silica zeolites from clear solution proceeds through amorphous nanoaggregates that gradually develop zeolite order in an apparently continuous amorphous to crystal transformation. Here, we combine molecular simulations with advanced algorithms that identify zeolite order and computer vision to elucidate the pathway from clear solution to zeolite nanocrystal. We show that at conditions of hydrothermal synthesis of silica zeolites, the transformation of precursor aggregate into zeolite is not limited by nucleation barriers but by the slow dynamics of reorganization in the glassy precursor matrix. The negligible nucleation barriers result in spinodal-like crystallization that leads to a gradual formation of a mosaic of small crystallites that explain the seemingly continuous character of zeolite crystallization and the catalytic activity of X-ray amorphous, protozeolites and embryonic zeolites. We find that zeolite-like porosity and short-range order emerge early within glassy precursors, well before crystallinity is detected in transmission electron microscopy (TEM) images or X-ray diffraction. The nanoaggregate’s temperature-size phase diagram reveals a convergence of the zeolite–amorphous equilibrium and maximum crystallization rate at ~3 nm diameter nanoparticle diameters and ~200 °C. This convergence signals the termination of the first-order amorphous-to-zeolite transition. Our results provide a unifying framework for understanding nucleation of silica zeolites from solution and suggest that barrierless nucleation may govern the formation of other nanoparticle systems, including minerals and oxides synthesized far below their bulk melting points.

36 MATERIALS SCIENCE↗
DOE OSTI2024

From Chaos to Clarity: Autonomous Materials Discovery for Extreme Environments [Slides]

The pursuit of advanced functional materials for energy applications demands an understanding of their behavior under the most challenging conditions. Extreme environments, characterized by intense radiation, high temperatures, and corrosive chemistries, push materials to their limits, often revealing unexpected behaviors and degradation pathways. Traditional materials research approaches, relying on trial-and-error experimentation, are often slow and resource-intensive, ill-suited to the complexities of extreme environments. This talk will explore the transformative potential of autonomous materials science in revolutionizing our understanding of materials synthesis and degradation in extreme environments. By integrating advanced microscopy techniques, artificial intelligence, and robotic experimentation, we can accelerate the discovery and design of resilient materials for a sustainable future. The presentation will highlight recent breakthroughs in autonomous microscopy, computer vision, and machine learning, showcasing their ability to unravel complex material transformations at the atomic scale. The talk will also delve into the challenges and opportunities associated with deploying autonomous systems to probe extreme environments, emphasizing the importance of robust algorithms, real-time data analysis, and adaptive experimentation. The ultimate goal is to empower scientists with unprecedented capabilities to explore, understand, and engineer materials that can withstand the harshest conditions, paving the way for innovations in energy, aerospace, and beyond.

14 SOLAR ENERGY↗