Search NASA⌕ Search

SEARCH · Search NASA

Results for “ViTs”

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 OSTI2023

A survey of techniques for optimizing transformer inference

Recent years have seen a phenomenal rise in the performance and applications of transformer neural networks. The family of transformer networks, including Bidirectional Encoder Representations from Transformer (BERT), Generative Pretrained Transformer (GPT) and Vision Transformer (ViT), have shown their effectiveness across Natural Language Processing (NLP) and Computer Vision (CV) domains. Transformer-based networks such as ChatGPT have impacted the lives of common men. However, the quest for high predictive performance has led to an exponential increase in transformers' memory and compute footprint. Researchers have proposed techniques to optimize transformer inference at all levels of abstraction. Further, this paper presents a comprehensive survey of techniques for optimizing the inference phase of transformer networks. We survey techniques such as knowledge distillation, pruning, quantization, neural architecture search and lightweight network design at the algorithmic level. We further review hardware-level optimization techniques and the design of novel hardware accelerators for transformers. We summarize the quantitative results on the number of parameters/FLOPs and the accuracy of several models/techniques to showcase the tradeoff exercised by them. We also outline future directions in this rapidly evolving field of research. We believe that this survey will educate both novice and seasoned researchers and also spark a plethora of research efforts in this field.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

A versatile machine learning workflow for high-throughput analysis of supported metal catalyst particles

Accurate and efficient characterization of nanoparticles (NPs), particularly regarding particle size distribution, is essential for advancing our understanding of their structure-property relationship and facilitating their design for various applications. In this study, we introduce a novel two-stage artificial intelligence (AI)-driven workflow for NP analysis that leverages prompt engineering techniques from state-of-the-art single-stage object detection and large-scale vision transformer (ViT) architectures. This methodology is applied to transmission electron microscopy (TEM) and scanning TEM (STEM) images of heterogeneous catalysts, enabling high-resolution, high-throughput analysis of particle size distributions for supported metal catalyst NPs. The model's performance in detecting and segmenting NPs is validated across diverse heterogeneous catalyst systems, including various metals (Ru, Cu, PtCo, and Pt), supports (silica (SiO 2 ), γ-alumina (γ-Al 2 O 3 ), and carbon black), and particle diameter size distributions with mean and standard deviations ranging from 1.6 ± 0.2 nm to 9.7 ± 4.6 nm. The proposed machine learning (ML) methodology achieved an average F1 overlap score of 0.91 ± 0.01 and demonstrated the ability to disentangle overlapping NPs anchored on catalytic support materials. The segmentation accuracy is further validated using the Hausdorff distance and robust Hausdorff distance metrics, with the 90th percent of the robust Hausdorff distance showing errors within 0.4 ± 0.1 nm to 1.4 ± 0.6 nm. In conclusion, our AI-assisted NP analysis workflow demonstrates robust generalization across diverse datasets and can be readily applied to similar NP segmentation tasks without requiring costly model retraining.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2025

Accelerating phase field simulations through a hybrid adaptive Fourier neural operator with U-net backbone

Prolonged contact between a corrosive liquid and metal alloys can cause progressive dealloying. For one such process as liquid-metal dealloying (LMD), phase field models have been developed to understand the mechanisms leading to complex morphologies. However, the LMD governing equations in these models often involve coupled non-linear partial differential equations (PDE), which are challenging to solve numerically. In particular, numerical stiffness in the PDEs requires an extremely refined time step size (on the order of 10 -12 s or smaller). This computational bottleneck is especially problematic when running LMD simulation until a late time horizon is required. This motivates the development of surrogate models capable of leaping forward in time, by skipping several consecutive time steps at-once. In this paper, we propose a U-shaped adaptive Fourier neural operator (U-AFNO), a machine learning (ML) based model inspired by recent advances in neural operator learning. U-AFNO employs U-Nets for extracting and reconstructing local features within the physical fields, and passes the latent space through a vision transformer (ViT) implemented in the Fourier space (AFNO). We use U-AFNOs to learn the dynamics of mapping the field at a current time step into a later time step. We also identify global quantities of interest (QoI) describing the corrosion process (e.g., the deformation of the liquid-metal interface, lost metal, etc.) and show that our proposed U-AFNO model is able to accurately predict the field dynamics, in spite of the chaotic nature of LMD. Most notably, our model reproduces the key microstructure statistics and QoIs with a level of accuracy on par with the high-fidelity numerical solver, while achieving a significant 11, 200 × speed-up on a high-resolution grid when comparing the computational expense per time step. Finally, we also investigate the opportunity of using hybrid simulations, in which we alternate forward leaps in time using the U-AFNO with high-fidelity time stepping. We demonstrate that while advantageous for some surrogate model design choices, our proposed U-AFNO model in fully auto-regressive settings consistently outperforms hybrid schemes.

36 MATERIALS SCIENCE↗
DOE OSTI2023

UVCGAN: UNet Vision Transformer cycle-consistent GAN for unpaired image-to-image translation

Unpaired image-to-image translation has broad applications in art, design, and scientific simulations. One early breakthrough was CycleGAN that emphasizes one-to-one mappings between two unpaired image domains via generative-adversarial networks (GAN) coupled with the cycle-consistency constraint, while more recent works promote one-to-many mapping to boost diversity of the translated images. Motivated by scientific simulation and one-to-one needs, this work revisits the classic CycleGAN framework and boosts its performance to outperform more contemporary models without relaxing the cycle-consistency constraint. To achieve this, we equip the generator with a Vision Transformer (ViT) and employ necessary training and regularization techniques. Compared to previous best-performing models, our model performs better and retains a strong correlation between the original and translated image. An accompanying ablation study shows that both the gradient penalty and self-supervised pre-training are crucial to the improvement. To promote reproducibility and open science, the source code, hyperparameter configurations, and pre-trained model are available at https: //github.com/LS4GAN/uvcgan.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

RingX: Scalable Parallel Attention for Long-Context Learning on HPC

The attention mechanism has become foundational for remarkable AI breakthroughs since the introduction of the Transformer, driving the demand for increasingly longer context to power frontier models such as large-scale reasoning language models and high-resolution image/video generators. However, its quadratic computational and memory complexities present substantial challenges. Current state-of-the-art parallel attention methods, such as ring attention, are widely adopted for long-context training but utilize a point-to-point communication strategy that fails to fully exploit the capabilities of modern HPC network architectures. In this work, we propose ringX, a scalable family of parallel attention methods optimized explicitly for HPC systems. By enhancing workload partitioning, refining communication patterns, and improving load balancing, ringX achieves up to 3.4 × speedup compared to conventional ring attention on the Frontier supercomputer. Optimized for both bi-directional and causal attention mechanisms, ringX demonstrates its effectiveness through training benchmarks of a Vision Transformer (ViT) on a climate dataset and a Generative Pre-Trained Transformer (GPT) model, Llama3 8B. Our method attains an end-to-end training speedup of approximately 1.5 × in both scenarios. To our knowledge, the achieved 38% model FLOPs utilization (MFU) for training Llama3 8B with a 1M-token sequence length on 4,096 GPUs represents one of the highest training efficiencies reported for long-context learning on HPC systems. Our code implementation is available at https://github.com/jqyin/ringX-attention.

Yin, Junqi [ORNL] (ORCID:0000000338435520)↗
DOE OSTI2025

ORBIT-2 Weather and Climate Downscaling Software Repository

ORBIT-2 is a scalable foundation model for global, hyper-resolution climate and weather downscaling. ORBIT-2 incorporates two key innovations: (1) Residual Slim ViT (Reslim), a lightweight architecture with residual learning and Bayesian regularization for efficient, robust prediction; and (2) TILES, a tile-wise sequence scaling algorithm that reduces self-attention complexity from quadratic to linear, enabling long-sequence processing and massive parallelism. ORBIT-2 scales to 10 billion parameters across 65,536 GPUs, achieving up to 4.1 ExaFLOPS sustained throughput and 74–98% strong scaling efficiency. It supports downscaling to 0.9 km global resolution and processes sequences up to 4.2 billion tokens. On 7 km resolution benchmarks, ORBIT-2 achieves high accuracy with 𝑅2 scores in range of 0.98–0.99 against observation data.

Wang, Xiao [Oak Ridge National Laboratory]↗
DOE OSTI2024

Vision Transformers Explained Series

Since their introduction in 2017 with Attention is All You Need¹, transformers have established themselves as the state of the art for natural language processing (NLP). In 2021, An Image is Worth 16x16 Words successfully adapted transformers for computer vision tasks. Since then, numerous transformer-based architectures have been proposed for computer vision. This article walks through the Vision Transformer (ViT) as laid out in An Image is Worth 16x16 Words.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

The VIBES Are Shifting: Assessing Emergent Capabilities in Multi-Modal Models

Researchers assessing open-source domains such as the internet, and particularly those studying information conflict, often have no single prescribed workflow. In the course of their research, they may need to perform a diverse array of tasks far beyond simply identifying an ever-changing set of objects. These analytical tasks can include ascertaining the provenance of images, understand an image in the context of accompanying text-based data, or identifying indicators of digital image manipulation. They must further be able to do this at the scale of tens of thousands of images or more. Traditional machine vision models have typically lacked the flexibility and breadth of performance sufficient for these needs. The research team from Pacific Northwest National Laboratory assessed the performance of a single baseline CLIP ViT-L model against a series of analytical tasks relevant for the study of online information conflict.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

A Semi-Supervised Learning Method for the Identification of Bad Exposures in Large Imaging Surveys

As the data volume of astronomical imaging surveys rapidly increases, traditional methods for image anomaly detection, such as visual inspection by human experts, are becoming impractical. We introduce a machine-learning-based approach to detect poor-quality exposures in large imaging surveys, with a focus on the DECam Legacy Survey (DECaLS) in regions of low extinction (i.e., E ( B − V ) < 0.04 ). Our semi-supervised pipeline integrates a vision transformer (ViT), trained via self-supervised learning (SSL), with a k-Nearest Neighbor (kNN) classifier. We train and validate our pipeline using a small set of labeled exposures observed by surveys with the Dark Energy Camera (DECam). A clustering-space analysis of where our pipeline places images labeled in good and bad categories suggests that our approach can efficiently and accurately determine the quality of exposures. Applied to new imaging being reduced for DECaLS Data Release 11, our pipeline identifies 780 problematic exposures, which we subsequently verify through visual inspection. Being highly efficient and adaptable, our method offers a scalable solution for quality control in other large imaging surveys.

Luo, Yufeng (ORCID:0000000246230683)↗
NASA NTRS1972

Dimensional-stability studies of candidate space-telescope mirror-substrate materials

The effects of aging, vacuum exposure, and thermal cycling on the dimensional stability of mirror-substrate materials, fused silica, Cer-Vit, Kanigen-coated beryllium, polycrystalline silicon, and U.L.E. fused silica were investigated. A multiple-beam interferometer was used to determine nonrecoverable surface-shape changes of the 12.7-cm-diameter mirrors with substrates of these materials. Thermal cycling and aging in vacuum produced the largest changes, but only a few were as large as 1/30 wavelength, where the wavelength was 632.8 nm.

Jerke, J. M.↗
NASA NTRS1973

Effects of surface polishing on the microstrain behavior of telescope mirror materials

Rough ground silicic mirror substrate materials were found in previous investigations to exhibit significant surface yield. This effect was removed by surface etching, a procedure not normally employed in the finishing of optical telescope mirrors. The effects of fine grinding and polishing techniques as well as graded etching are investigated. Torsional shear measurements of yield strain versus stress are made on four candidate mirror substrate materials: polycrystalline silicon, ULE silica 7971, CER-VIT 101, and fused silica 7940. Commonly employed fine grinding and polishing practices are shown to remove a major portion of the surface yield found in rough ground mirror substrate materials.

Eul, W. A.↗
NASA NTRS1972

Indium sealing techniques.

Gold films are used as an alloying flux to form 5-micron-thick indium film seals at temperatures below 300 C. Pyrex was sealed to quartz, ULE, CER-VIT, Irtran 2, Ge, GaAs, Invar, Kovar, Al, and Cu. The seals can also be used as current feedthroughs and graded seals.

Hochuli, U.↗
NASA NTRS1973

Temperature compensated sleeve type mirror mount

The primary mirror of a large (26-inch diameter aperture) solar telescope was made of glass ceramic and designed with an integral hub on the back of the center of the mirror. This permits heat from the mirror to radiate off its back to a nearby cold plate. To permit mounting without high stresses, the hub was ground down to a smooth cylindrical surface 3.5 inch in diameter. The ground surface was then acid-etched to remove 0.007 inch (on the diameter) by immersion for five minutes in a mixture of four parts 92% sulfuric acid and three parts 50% hydrofluoric acid. The acid etching removes microcracks from the ground Cer-Vit surface. An Invar sleeve was fabricated to fit over the hub with about 0.010 inch radial (0.020 inch diametral) clearance.

Source record↗
NASA NTRS1974

The 1974 NASA-ASEE summer faculty fellowship aeronautics and space research program

Research activities by participants in the fellowship program are documented, and include such topics as: (1) multispectral imagery for detecting southern pine beetle infestations; (2) trajectory optimization techniques for low thrust vehicles; (3) concentration characteristics of a fresnel solar strip reflection concentrator; (4) calaboration and reduction of video camera data; (5) fracture mechanics of Cer-Vit glass-ceramic; (6) space shuttle external propellant tank prelaunch heat transfer; (7) holographic interferometric fringes; and (8) atmospheric wind and stress profiles in a two-dimensional internal boundary layer.

Obrien, J. F., Jr.↗
NASA NTRS1975

Measurement of dimensional stability

A technique was developed for measuring, with a precision of one part 10 to the 9th power, changes in physical dimensions delta L/L. Measurements have commenced on five materials: Heraeus-Schott Homosil (vitreous silica), Corning 7940 (vitreous silica), Corning ULE 7971 (titanium silicate), Schott Zero-Dur, and Owens-Illinois Cer-Vit C-101. The study was extended to include Universal Cyclops Invar LR-35 and Simonds-Saw Superinvar.

Jacobs, S. F.↗