SEARCH · Search NASA

Results for “Foundation models”

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.

SourceResearch connectionFrom yearThrough year

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

At least 127 records · Page 7

All sources NASA NTRS DOE OSTI

DOE OSTI2024

Optimizing Distributed Training on Frontier for Large Language Models

Large language models (LLMs) have demonstrated remarkable success as foundational models, benefiting various downstream applications through fine-tuning. Loss scaling studies have demonstrated the superior performance of larger LLMs compared to their smaller counterparts. Nevertheless, training LLMs with billions of parameters poses significant challenges and requires considerable computational resources. For example, training a one trillion parameter GPT-style model on 20 trillion tokens requires a staggering 120 million exaflops. This research explores efficient distributed training strategies to extract this computation from Frontier, the world's first exascale supercomputer. We enable and investigate various model and data parallel training techniques, such as tensor parallelism, pipeline parallelism, and sharded data parallelism, to facilitate training a trillion-parameter model on Frontier. We empirically assess these techniques and their associated parameters to determine their impact on memory footprint, communication latency, and GPU's computational efficiency. We analyze the complex interplay among these techniques and find a strategy to combine them to achieve high throughput through hyperparameter tuning. We have identified efficient strategies for training large LLMs of varying sizes through empirical analysis and hyperparameter tuning. For 22 Billion, 175 Billion, and 1 Trillion parameters, we achieved GPU throughputs of 38.38%, 36.14%, and 31.96%, respectively. For the training of the 175 Billion parameter model and the 1 Trillion parameter model, we achieved 100% weak scaling efficiency on 1024 and 3072 Mi250X GPUs, respectively. We also achieved strong scaling efficiencies of 89% and 87% for these two models. We trained these models only tens of iterations instead of training till completion.

Yin, Junqi↗

DOE OSTI2025

Exocortex Network for AI-Augmented Human-Led Scientific Expedition

AI advances in science can be viewed along two main directions with a fluid boundary: enhancing efficiency through automation and smart tools to accelerate tasks that humans can already perform; and enabling exploration into uncharted territories and potentially toward AGI. These advances manifest in the AI cognitive core through the development and explainability of foundation models; in the physical embodiment of instruments and facilities; and in the integrated agency of AI workflows exemplified by the science exocortex. To address the role of humans in this evolving landscape, in this Perspective, we suggest a third direction: the development of personalized agents that form human-centered networks, supporting both efficiency and exploration while ensuring that AI remains aligned with human vision.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

Towards the next generation of Geospatial Artificial Intelligence

Geospatial Artificial Intelligence (GeoAI), as the integration of geospatial studies and AI, has become one of the fastest-developing research directions in spatial data science and geography. This rapid change in the field calls for a deeper understanding of the recent developments and envision where the field is going in the near future. In this work, we provide a quantitative analysis of the GeoAI literature from the spatial, temporal, and semantic aspects. We briefly discuss the history of AI and GeoAI by highlighting some pioneering work. Then we discuss the current landscape of GeoAI by selecting five representative subdomains including remote sensing, urban computing, Earth system science, cartography, and geospatial semantics. Finally, we highlight several unique future research directions of GeoAI which are classified into two groups: GeoAI method development challenges and GeoAI Ethics challenges. Topics include heterogeneity-aware GeoAI, knowledge-guided GeoAI, spatial representation learning, geo-foundation models, fairness-aware GeoAI, privacy-aware GeoAI, as well as interpretable and explainable GeoAI. We hope our review of GeoAI’s past, present, and future is comprehensive and can enlighten the next generation of GeoAI research.

58 GEOSCIENCES↗

DOE OSTI2025

Spatiotemporal forecasting of the edge localized modes in tokamak plasmas using neural networks

Artificial intelligence techniques have been increasingly adopted by the plasma and fusion science to address problems like plasma reconstruction, surrogate modeling, and tokamak/stellarator optimization. A key focus in sustained fusion research is the prediction and mitigation of edge-localized-modes (ELMs), instabilities that occur in short, periodic bursts and can cause erosion to the tokamak vessel wall. Recent research has demonstrated the power of neural networks in approximating continuous functions. In this work, we build spatiotemporal forecasting models that can predict the onset of ELMs and their evolution at early stages. We leverage recent advances in generative modeling, sequence-to-sequence modeling, and Fourier neural operators to propose architectures and training strategies that can learn to forecast short to long term dynamics of the noisy signals due to ELMs. We benchmark the developed model against a state-of-the-art foundation model using the beam emission spectroscopy (BES) data that captures the plasma fluctuations due to ELMs over a 8 x 8 spatial grid. Our models demonstrate high accuracy, outperforming the baselines, in predicting the evolution of BES signals during ELM events. Furthermore, the developed models exhibit high accuracy in predicting the rapid rise and relaxation of the signals due to ELMs within 30–80 µs.

edge localized modes↗

DOE OSTI2024

Segmentation Model Distillation [Poster]

The process of training object detection (OD) or image segmentation model requires both a substantial amount of data and technical knowledge, which often creates challenges in applying these types of models to their full potential. In order to streamline the process of developing these models, we propose a new pipeline where a foundation model assists in the dataset generation. Then this resulting dataset is used to fine-tune a fast light-weight model to perform the custom segmentation or OD. This resulting model is also fit for real-time image segmentation, such as in a video stream.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

High-Fidelity CFD Modeling of Cryogenic Hydrogen Isotope Extrusion for Fusion Reactor Pellet Fueling

This study investigates the extrusion processes of deuterium and protium using ANSYS-Polyflow. The geometries and computational fluid dynamics (CFD) settings closely replicate the experimental setups and data acquired from the extruder experiments at Oak Ridge National Laboratory (ORNL) for validation purposes. We explore the impacts of (1) slip versus non-slip boundary conditions and (2) the use of constant, temperature-, and shear rate–dependent viscosities, concluding that the implementation of non-slip wall boundary conditions combined with shear rate–dependent viscosity produced more accurate predictions. The simulations achieved excellent agreement with the experimental data, with relative differences of only 5% for deuterium, and 3% to 6% for protium. This is the first time that experimental extrusion data at ORNL have been accurately predicted through high-fidelity CFD modeling. In conclusion, the advancements offer valuable insights and a foundational modeling tool for optimizing pellet injectors for ITER and other future reactor-scale devices.

ANSYS-Polyflow↗

DOE OSTI2023

BuildingsBench: A Benchmark for Universal Building Load Forecasting [SWR-23-51]

The residential and commercial building stock in the United States is responsible for a significant percentage of energy consumption and greenhouse gas emissions. Electrification of end-uses, as well as decarbonizing the electrical grid through renewable energy sources such as solar and wind, constitutes the pathway to zero-emission buildings. Forecasting day-ahead building energy consumption is an integral part of this solution. Currently, specialized forecasting models are hand-made for each individual building, which is time-consuming, expensive, and leads to duplicated efforts. BuildingsBench is a Python software framework for training and comparing generalized machine learning models for universal building load forecasting. This challenge tasks a single foundational model to generalize its forecasts for a wide variety of buildings, across geographic regions, building types, weather patterns, and more. This software provide code for pre-training such models and subsequently evaluating their performance on a suite of hundreds of diverse real and synthetic buildings. BuildingsBench is a platform for: - Large-scale pretraining with the synthetic Buildings-900K dataset for short-term load forecasting (STLF). Buildings-900K is statistically representative of the entire U.S. building stock and is extracted from the NREL End-Use Load Profiles database. - Benchmarking on two tasks evaluating generalization: zero-shot STLF and transfer learning for STLF. We provide an index-based PyTorch Dataset for large-scale pretraining, easy data loading for multiple real building energy consumption datasets as PyTorch Tensors or Pandas DataFrames, simple (persistence) to advanced (transformer) baselines, metrics management, and more.

Emami, Patrick↗

DOE OSTI2023

Particle resuspension: Challenges and perspectives for future models

Using what has become a celebrated catchphrase, Philip W. Anderson once wrote that “more is different” (Science, Vol. 177, Issue 4047, pp. 393–396, 1972). First formulated in the context of condensed matter, this statement carries far beyond the sole limits of solid-state physics. It emphasizes that collective behavior can be more than the mere sum of what happens for elementary constituents or the mere collation of the evolution of each degree of freedom. Said otherwise, complex phenomena can arise out of the interplay between multiple sub-phenomena each of which can be relatively simple. The process of particle resuspension, in which discrete particles adhering on a surface are pulled off and carried away by a fluid flow, is another example involving a web of phenomena pertaining to fluid mechanics, particle dynamics and interface chemistry whose cross-effects create an intricate topic. The purpose of this review is to analyze the physics at play in particle resuspension in order to bring insights into the rich complexity of this common but challenging concern. Following the more-is-different vision, this is performed by starting from a range of practical observations and experimental data. We then work our way through the investigation of the key mechanisms which play a role in the overall process. In turn, these mechanisms reveal an array of fundamental interactions, such as particle–fluid, particle–particle and particle–surface, whose combined effects create the tapestry of current applications. At the core of this analysis are descriptions of these physical phenomena and the different ways through which they are intertwined to build up various models used to provide quantitative assessment of particle resuspension. The physics of particle resuspension implies to hold together processes occurring at extremely different space and time scales and models are key in providing a single vehicle to lead us through such multiscale journeys. This raises questions on what makes up a model and one objective of the present work is to clarify the essence of a modeling approach. In spite of its ubiquitous nature, particle resuspension is still at the early stages of developments. Many extensions need to be worked out and revisiting the art of modeling is not a moot point. The need to consider more complex objects than small and spherical particles and, moreover, to come up with unified descriptions of mono- and multilayer resuspension put the emphasis on solid model foundations if we are to go beyond current limits. Further, this is very much modeling in the making and new ideas are proposed to stimulate interest into this everyday but challenging issue in physics.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

NASA NTRS2016

Canonical Statistical Model for Maximum Expected Immission of Wire Conductor in an Aperture Enclosure

Prediction of the maximum expected electromagnetic pick-up of conductors inside a realistic shielding enclosure is an important canonical problem for system-level EMC design of space craft, launch vehicles, aircraft and automobiles. This paper introduces a simple statistical power balance model for prediction of the maximum expected current in a wire conductor inside an aperture enclosure. It calculates both the statistical mean and variance of the immission from the physical design parameters of the problem. Familiar probability density functions can then be used to predict the maximum expected immission for deign purposes. The statistical power balance model requires minimal EMC design information and solves orders of magnitude faster than existing numerical models, making it ultimately viable for scaled-up, full system-level modeling. Both experimental test results and full wave simulation results are used to validate the foundational model.

Computational electromagnetics↗

DOE OSTI2025

Automatic building energy model development and debugging using large language models agentic workflow

Building energy modeling (BEM) is a complex process that demands significant time and expertise, limiting its broader application in building design and operations. While Large Language Models (LLMs) agentic workflow have facilitated complex engineering processes, their application in BEM has not been specifically explored. This paper investigates the feasibility of automating BEM using LLM agentic workflow. Here, we developed a generic LLM-planning-based workflow that takes a building description as input and generates an error-free EnergyPlus building energy model. Our robust workflow includes four core agents: 1) Building Description Pre-Processing, 2) IDF Object Information Extraction, 3) Single IDF Object Generator Suite, and 4) IDF Debugging Agent. These agents divide the complex tasks into manageable sub-steps, enabling LLMs to generate accurate and reliable results at each stage. The case study demonstrates the successful translation of a building description into an error-free EnergyPlus model for the iUnit modular building at the National Renewable Energy Laboratory. The effectiveness of our workflow surpasses: 1) naive prompt engineering, 2) other LLM-based workflows, and 3) manual modeling, in terms of accuracy, reliability, and time efficiency. The paper concludes with a discussion on the interplay between foundational models and LLM agent planning design, advocating for the use of fine-tuned, specialized models to advance this field.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2026

Particle trajectory representation learning with masked point modeling

Liquid argon time projection chambers (LArTPCs) offer millimeter-scale 3D images of particle trajectories, enabling precision studies of neutrino oscillation, detection of supernova and solar neutrinos, searches for exotic dark matter, and proton decay. Current approaches utilize supervised machine learning models, requiring extensive simulations of particle physics and detector response that can introduce bias. Self-supervised learning (SSL), a machine learning approach that learns useful representations of unlabeled data from the data itself, has significantly advanced how large datasets are utilized for representation learning; however, its potential for applications to sensory data in high precision particle physics experiments remains largely unexplored. We introduce the Point-based liquid argon masked autoencoder (PoLAr-MAE), a self-supervised framework that learns physically meaningful representations directly from unlabeled LArTPC images. PoLAr-MAE achieves remarkable data efficiency for a point-level segmentation task, outperforming fully supervised methods in low data regimes. Linear classifiers on model outputs demonstrate robust performance across multiple downstream tasks. Our results position sensor-level SSL as a practical foundation model strategy for LArTPCs.

Young, Samuel [Stanford Univ., CA (United States)]↗

DOE OSTI2025

Method to simultaneously facilitate all jet physics tasks

Machine learning has become an essential tool in jet physics. Due to their complex, high-dimensional nature, jets can be explored holistically by neural networks in ways that are not possible manually. However, innovations in all areas of jet physics are proceeding in parallel. We show that specially constructed machine learning models trained for a specific jet classification task can improve the accuracy, precision, or speed of all other jet physics tasks. This is demonstrated by training on a particular multiclass generation and classification task and then using the learned representation for different generation and classification tasks, for datasets with a different (full) detector simulation, for jets from a different collision system ($pp$ versus $ep$), for generative models, for likelihood ratio estimation, and for anomaly detection. We consider our omnilearn approach thus as a jet-physics foundation model. It is made publicly available for use in any area where state-of-the-art precision is required for analyses involving jets and their substructure.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

NASA NTRSDate not supplied

Improving Sim-to-Real Transfer in Vision-Based Robot Navigation Via Instance-Level GAN-Based Data Augmentation

Achieving robust vision-based robotic tasks requires large amounts of data, which are often difficult to obtain in real-world scenarios. Simulators and synthetic data offer a cost-effective alternative, but the visual gap between simulation and reality hinders the performance of models when deployed in real-world environments. In this paper, we present a data augmentation pipeline that integrates a foundation model (Segment Anything Model) with an unsupervised image-to-image translation model (CycleGAN) for instance-level domain transfer from simulation to reality. This pipeline enables the generation of realistic labeled data from synthetic images for training supervised machine learning models in vision-based navigation tasks. We evaluate our approach on real-world data for ego-vehicle pose estimation, a critical autonomous navigation task involving the prediction of cross-track position and heading angle relative to road center line markings. The results of our tests show that our GAN-based data augmentation pipeline significantly outperforms models trained solely on simulation data or on data processed with standard image augmentation methods for sim-to-real transfer, enhancing model robustness and generalizability in real-world scenarios. Our method provides a scalable and flexible data augmentation tool for leveraging large synthetic datasets to enhance vision-based robotic navigation tasks.

artificial intelligence↗

DOE OSTI2024

SCITUNE: Aligning Large Language Models with Human-Curated Scientific Multimodal Instructions

Instruction finetuning is a popular paradigm to align large language models (LLM) with human intent. Despite its popularity, this idea is less explored in improving the LLMs to align existing foundation models with scientific disciplines, concepts and goals. In this work, we present SciTune as a tuning framework to improve the ability of LLMs to follow scientific multimodal instructions. To test our methodology, we use a human-generated scientific instruction tuning dataset and train a large multimodal model LLaMA-SciTune that connects a vision encoder and LLM for science-focused visual and language understanding. LLaMA-SciTune significantly outperforms the state-of-the-art models in the generated figure types and captions in multiple scientific multimodal benchmarks. In comparison to the models that are fine-tuned with machine generated data only, LLaMA-SciTune surpasses human performance on average and in many sub-categories on the ScienceQA benchmark.

• Artificial intelligence (AI) / machine learning ↗

DOE OSTI2025

Location generalizability of image-based air quality models

This paper is to be submitted at the IEEE/CVF Winter Conference on Applications of Computer Vision (WACV) Computer Vision for Earth Observation workshop. The full paper abstract is below: The ability to rapidly quantify atmospheric pollutants is important both for global emissions monitoring and for mitigating the adverse effects that follow a hazardous chemical release. In the aftermath of a chemical release, imagery is often the only available resource to assess local conditions. Recent work has demonstrated initial success in predicting particulate matter pollution from imagery; however, these results are tied to a specific site and do not generalize to new geographic locations. In this work, we seek to understand how easily deep learning models generalize to new locations in the context of image-based air quality assessments, targeting two distinct tasks: (1) broad measures of particulate matter pollution, and (2) the mass of a given chemical released in hazardous plumes. For the latter, we focus on sulfur dioxide, a toxic aerosol and a major component of particulate matter pollution caused by industrial fossil fuel consumption. To develop a model that operates in the widest possible range of environments, we test different training strategies, including the use of new geolocation foundation models. The best performing models achieve >80% accuracy when evaluating unseen imagery at previously seen sites, but we find significant drops in performance when evaluating imagery from unseen sites, at best 65%. Additionally, we present the public release of the National Parks Air Quality Index Dataset, a new medium-sized dataset that pairs imagery with sensor-based air quality measurements at 15 different national parks.

Byler, Eleanor B. [BATTELLE (PACIFIC NW LAB)]↗

DOE OSTI2025

Harnessing distributed GPU computing for generalizable graph convolutional networks in power grid reliability assessments

Although machine learning (ML) has emerged as a powerful tool for rapidly assessing grid contingencies, prior studies have largely considered a static grid topology in their analyses. This limits their application, since they need to be re-trained for every new topology. Here, this paper explores the development of generalizable graph convolutional network (GCN) models by pre-training them across a range of grid topologies and contingency types. We found that a GCN model with auto-regressive moving average (ARMA) layers with a line graph representation of the grid offered the best predictive performance in predicting voltage magnitudes (VM) and voltage angles (VA). We introduced the concept of phantom nodes to consider disparate grid topologies with a varying number of nodes and lines. For pre-training the GCN ARMA model across a variety of topologies, distributed graphics processing unit (GPU) computing afforded us significant training scalability. The predictive performance of this model on grid topologies that were part of the training data is substantially better than the direct current (DC) approximation. Although direct application of the pre-trained model to topologies that are not part of the grid is not particularly satisfactory, fine-tuning with small amounts of data from a specific topology of interest significantly improves predictive performance. In general, this paper highlights the feasibility of training large-scale GNN models to assess the reliability of power grids by considering a wide variety of grid topologies and contingency types. With the advent of foundational models in ML and the exponential increase in GPU computing clusters, generalizable ML models will significantly enhance how utilities manage power systems and make decisions in real-time or near-real-time.

24 - POWER TRANSMISSION AND DISTRIBUTION↗

DOE OSTI2026

Generative AI for Grid Operations [Slides]

In the last few years, the development and use of generative artificial intelligence (AI) and large-language models (LLMs) have changed the landscape of how AI and machine learning (ML) are being used in power systems. LLMs are built on foundational models based on large data sets that can be trained to provide information rapidly and through simple natural language prompts. Generative AI can then perform human-like tasks using ML models to identify and mimic pattens in the data sets. This presentation explores how generative AI can enhance grid operations by improving forecasts, enabling rapid contingency analyses, and offering real-time operational suggestions. By providing grid operators with valuable insights, generative AI will empower them to manage power systems more effectively.

24 POWER TRANSMISSION AND DISTRIBUTION↗

NASA NTRS1995

Report on twisted nematic and supertwisted nematic device characterization program

In this study we measured the optical characteristics of normally white twisted nematic (NWTN) and super twisted nematic (STN ) cells. Though no dynamic computer model was available, the static observations were compared with computer simulated behavior. The measurements were taken as a function of both viewing angle and applied voltage and included in the static case not only luminance but also contrast ratio and chromaticity . We employed the computer model Twist Cell Optics, developed at Kent State in conjunction with this study, and whose optical modeling foundation, Iike the ViDEOS program, is the 4 x 4 matrix method of Berreman. In order to resolve discrepancies between the experimental and modeled data the optical parameters of the individual cell components, where not known, were determined using refractometry, profilometry, and various forms of ellipsometry. The resulting agreement between experiment and model is quite good due primarily to a better understanding of the structure and optics of dichroic sheet polarizers. A description of the model and test cells employed are given in section 2. Section 3 contains the experimental data gathered and section 4 gives examples of the fit between model and experiment. Also included with this report are a pair of papers which resulted from the research and which detail the polarizer properties and some of the cell characterization methods.

Source record↗