SEARCH · Search NASA
Results for “foundation model”
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.
Zatom v1.0.0
Zatom is an AI foundation model for 3D molecules and materials.
Reasoning4PDEs
Towards Reasoning for PDE Foundation Models: A Reward-Model-Driven Inference-Time-Scaling Algorithm
HydraGNN_GFM_FineTuning4Materials v1.0
This repository enables fine-tuning of the HydraGNN Predictive GFM 2026 — an open-source ensemble of pre-trained graph foundation models for atomistic materials modeling, developed at Oak Ridge National Laboratory. The GFM 2026 is freely available and downloadable via Globus from the OLCF Data Constellation (DOI: 10.13139/OLCF/2562660). Starting from these pre-trained weights, this repository provides a complete transfer learning pipeline for adapting the GFM ensemble to domain-specific molecular and materials property prediction tasks. It includes: 1) Utilities for ensemble fine-tuning with task-specific output heads 2) Example pipelines for eight widely-used materials and molecular datasets 3) Tools for model adaptation and head configuration 4) Data preprocessing utilities for each supported dataset 5) Benchmarking and evaluation scripts
Sequence length scaling in vision transformers for scientific images on frontier
Vision Transformers (ViTs) are pivotal for foundational models in scientific imagery, including Earth science applications, due to their capability to process large sequence lengths. While transformers for text have inspired scaling sequence lengths in ViTs, adapting these for ViTs introduces unique challenges. We develop distributed sequence parallelism for ViTs, enabling them to handle up to 1M tokens. Our approach, leveraging DeepSpeed-Ulysses and Long-Sequence-Segmentation with model sharding, is the first to apply sequence parallelism in ViT training, achieving a 94% batch scaling efficiency on 2,048 AMD-MI250X GPUs. Evaluating sequence parallelism in ViTs, particularly in models up to 10B parameters, highlighted substantial bottlenecks. We countered these with hybrid sequence, pipeline, and flash attention strategies, to scale beyond single GPU memory limits. Our method significantly enhances climate modeling accuracy by 20% in temperature predictions, marking the first training of a vision transformer model to convergence with a sequence length of 188K tokens, using full self-attention.
2025 Workshop on Envisioning Frontiers in AI and Computing for Biological Research: Position Papers
This workshop aims to identify key research directions for transforming biology using artificial intelligence (AI), machine learning (ML) and computational methods to facilitate the discovery of new behaviors, mechanisms, and designs of biological processes relevant to DOE missions, underpinning a broader U.S. bioeconomy. By developing novel AI/ML technologies to analyze and interpret complex biological data, researchers can organize and simulate biological processes at various scales as well as advance predictive understanding and manipulation of biological systems. This integration of computation, experimentation, and next-generation experimental technologies can lead to discoveries in new biological behaviors and mechanisms relevant to DOE missions. The focus is on how advanced computational and mathematical methods can impact this mission by exploring digital twins, foundation models, automated laboratory experiments, modeling of complex living systems, and data-driven approaches for the biodesign of plants and microbial systems. While data management is important, it is not the primary focus of this workshop, which will assess the current state, trends, and AI/ML challenges at the interface between biology and computational science to identify opportunities for high-impact research at their intersection. The goal is to define research needs and opportunities that align with biological sciences, computational sciences, and applied mathematics research.
Brochure for the DOE Office of Science Workshop on Envisioning Frontiers in AI and Computing for Biological Research
In February of 2025 a joint ASCR/BER workshop was held to identify key transformational research directions for understanding biology using artificial intelligence (AI), digital twins and high-performance (HPC) computational methods to facilitate scientific discovery and innovation in support of the Department of Energy mission. AI technologies offer exciting new groundbreaking methods to analyze large volumes of complex biological data, thereby greatly accelerating the ability to understand, predict, and design biological processes for beneficial purposes. In the laboratory, the bridging of AI-enabled automated experimental technologies, HPC and digital twins will provide potent tools for researchers to explore the fundamental nature of biology and harness its inherent metabolic potential for a variety of beneficial purposes. The focus of this workshop was on how high-performance computational methods can impact this objective by exploring digital twins, foundational models, and data-driven approaches with applications to advance automated laboratory experiments, modeling of complex living systems and engineering new functions into plants and microbial systems relevant to DOE mission. Workshop attendees with expertise in plant science, microbiology, mathematics, computer science, and AI assessed the current state of the science, trends, and AI challenges at the interface of plant and microbial systems biology and computational science to identify opportunities for high-impact research. This collaborative effort capitalized on ASCR's advancements in applied mathematics, computer science, and Exascale systems, and BER's expertise in basic genomics-enabled research on DOE relevant plant and microbial systems. The workshop culminated in four key priority research directions to guide future research and development within DOE Office of Science programs.
Electron-Proton Scattering Event Generation using Structured Tokenization
Recent work such as Omnijet-$\alpha$ has demonstrated that effective tokenization combined with transformer-based architectures can produce effective foundation models for jet physics. While tokenization may help models capture generalizable event characteristics, it also introduces discretization errors that may compromise the precision required for downstream physics analyses. As the number and complexity of the particle features grow, these errors are likely to grow proportionally. In this study, we investigate new tokenization strategies to improve the application of generative transformer models to \textsc{Pythia8} simulations of electron-proton scattering at the Electron-Ion Collider. Specifically, we propose a feature-based structured tokenization approach that utilizes multiple tokens per particle, improving expressivity, while reducing the total number of unique tokens needed. We evaluate this method against grid-based binning, K-means clustering, and vector-quantized variational auto-encoders on the event simulations. Our results show that feature-based structured tokenization reduces discretization error, leading to more accurate generative modeling of particle-level events.
Structure-Aware Unsupervised, Transformational Machine Learning for Drug Discovery (DTRA Basic Research Final Report)
The major goal of this project is to develop machine learning (ML) methods to enable improved predictive power on real drug discovery for novel targets. More specifically, we planned to demonstrate the capability and effectiveness of ML tools utilizing unlabeled large-volume protein-ligand datasets. We investigated multiple pre-training approaches for 3D protein-ligand structure-based foundation models, without relying on experimental binding data. We also addressed scenarios in which crystal structures are unavailable or binding data are limited. We also planned to develop a complete pipeline to screen novel compounds as well as to demonstrate the capability and effectiveness of the developed methods by testing on a realistic drug discovery task such as SARS-CoV-2. While the major goals and milestones remain consistent with the original proposal, certain technical details have been adjusted, based on the experimental results and related outcomes.
Toward Trustworthy Autonomous Science: A Two-Year Community Roadmap
One year ago, the AISLE roadmap argued that autonomous laboratories operated as isolated islands and proposed a grassroots network organized around five critical dimensions. The field has since moved faster than that roadmap anticipated: multi-agent systems have produced experimentally validated hypotheses, self-driving laboratories have grown more interoperable and orchestrated, reasoning-trained and domain foundation models have raised the capability ceiling, and the Genesis Mission has placed autonomous experimentation at the center of U.S. federal science strategy, with industry emerging as a primary actor. Progress has met a sobering counter-current, including a corrected flagship discovery result, benchmarks showing that agents which rival experts on closed-ended questions still complete only a fraction of open-ended research, and fabricated citations surfacing at leading venues. We read this as the defining tension of the field: producing a candidate discovery is no longer the hard part, but verifying it is, and this asymmetry now limits autonomous science more than raw model capability. Accordingly, we update the roadmap around seven dimensions, revisiting the original five and elevating two former cross-cutting concerns, trust, verification, and reproducibility, and safety, security, and governance, to first-class status. We assess the original milestones (M1 through M14) as achieved, partially achieved, reframed, or open, add four new milestones (M15 through M18) for the elevated dimensions, and scope the path forward to a two-year horizon, with the first year concentrating on interfaces, protocol adoption, and the scaffolding of verification, and the second targeting federation, zero-trust coordination, and governance. Throughout, we position the grassroots network as the interoperability fabric that lets national programs, international initiatives, and commercial platforms connect rather than re-silo.
Cloverleaf Data Artifacts for ArtIMis LDRD
This report summarizes the use of the open-source CloverLeaf/CloverLeaf3D mini-apps to generate synthetic data sets to train foundation models for the ArtIMis LDRD DI. These data artifacts are intended to be used by LANL collaborators and shared externally with our university and institutional partners. Note that CloverLeaf/CloverLeaf3D is not a LANL simulation code.
Advancing Open Source Science Initiatives Through Public-Private Partnerships
Collaboration is fundamental to advancing open science within the science community. With the recent developments in technology and research, the establishment of formal partnerships between the private sector and government agencies are needed to bridge the knowledge gaps and expedite the time to actionable science. NASA’s Interagency Implementation and Advanced Concepts Team (IMPACT) seeks to address this challenge by establishing non-reimbursable Space Act Agreements with industry leaders in cloud computing, artificial intelligence (AI) and machine learning. The purpose of these agreements is to advance open source science initiatives in the areas of data discovery, access and use of high value NASA science data sets on the cloud. As well as, jointly work on common research problems to accelerate the development and adoption of new AI technologies. Current success stories include co-locating NASA datasets from multiple science disciplines on one platform using Amazon Web Services Open Data Registry, developing AI Foundation Models for Science with IBM and co-hosting training workshops and tutorials for the science community aimed at providing hands-on experience with using NASA data and AI models on the cloud. In summary, we will present an overview of our partnerships supporting open source science initiatives, describe current activities and lessons learned that may be useful to others considering similar partnerships with the private sector.
Artificial Intelligence Workshop Report
The 4th NASA Science Mission Directorate (SMD) Artificial Intelligence (AI) Workshop, held during March 25-27, 2024, in Huntsville, AL, highlighted the significant potential of AI and machine learning (ML) in scientific research and processes. The workshop, supported by the NASA Office of Chief Science Data Officer (OCSDO), emphasized the critical role of foundation models (FMs) and large language models (LLMs) in advancing scientific disciplines. The event brought together domain scientists, computer scientists, AI experts, program managers, program scientists, and industry partners to address key challenges and explore opportunities in applying these advanced technologies.
Advancing Wildfire Monitoring with TEMPO and ML tools: Hourly Smoke and Fire‑Front Mapping and Near‑Surface NO₂ Predictions
Wildfires impose substantial impacts on communities and regions downwind of wildfire smoke. We present a TEMPO‑enabled workflow that generates value‑added Level‑3 smoke‑plume masks and fire‑front maps for large wildfires, such as 2024 Park Fire, using the self‑supervised deep learning system SIT‑FUSE, along with near‑surface NO₂ predictions produced by a foundation model (Microsoft Aurora). We conclude by outlining a roadmap for expanding these capabilities to additional Western U.S. wildfire events and for delivering actionable tools to stakeholders. This open-source, reproducible workflow provides a scalable framework for cross-agency wildfire monitoring to overcome traditional limitations in smoke-cloud discrimination and air-quality forecasting by incorporating TEMPO data and beyond.
Development of New Reactor Core Configuration for Power Uprate - Fuel Reload & Heat Processing Analyses, Core Design, System Safety Assessments, and Fuel Performance Analyses
With the passage of the Infrastructure Investment and Jobs Act in 2021 and the Inflation Reduction Act (IRA) in 2022, the United States stands at a critical juncture for the future of nuclear power. These landmark policies provide significant support for clean energy initiatives, positioning nuclear power as a key component of the nation’s strategy to reduce carbon emissions and achieve energy security. This growing emphasis on nuclear energy is driven by the need for reliable, low-carbon power sources as the country transitions away from fossil fuels. Federal policy, along with increasing state-level support, is encouraging investment in nuclear technology advancements to meet these demands. Building new nuclear power plants (NPPs), however, presents significant challenges due to high costs and long construction timelines. As a result, increasing the power output of existing NPPs through power uprates has emerged as a more feasible and cost-effective strategy. One key area of advancement is the development of accident-tolerant fuel (ATF), such as chromium-coated zirconium alloy cladding, which offers enhanced material performance, enabling power uprates in light water reactors (LWRs). Given the growing demand for nuclear energy fueled by federal policies and state initiatives, it is essential to evaluate the feasibility and benefits of significant power uprates in existing pressurized water reactors (PWRs) using advanced fuel technologies. The introduction of ATF concepts opens new opportunities for safely and economically achieving these power increases. Assessing whether these innovations can support substantial power uprates while maintaining operational safety is crucial to maximizing the potential of the nation’s existing nuclear infrastructure. This project aims to explore how power uprates can be achieved by boosting reactor thermal power output and optimizing reactor core design, while ensuring the safety and economic viability of NPPs. Specifically, it will focus on demonstrating the technical and economic feasibility of power uprates in a PWR using low 5-10% enrichment uranium (LEU+) high burnup (HBU) fuel combined with ATF concepts. In fiscal year 2024 (FY24), the research and development focus on building foundational models and conducting multi-physics performance and safety analyses to support the power uprate. The findings of the study would be shared through LWRS Seasonal Meetings, conferences and workshops with utility companies and researchers. These also serve as a basis for further study of fuel reloading optimization with ATF claddings.
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.
Electron-Proton Scattering Event Generation using Structured Tokenization
Recent work such as Omnijet-$\alpha$ has demonstrated that effective tokenization combined with transformer-based architectures can produce effective foundation models for jet physics. While tokenization may help models capture generalizable event characteristics, it also introduces discretization errors that may compromise the precision required for downstream physics analyses. As the number and complexity of the particle features grow, these errors are likely to grow proportionally. In this study, we investigate new tokenization strategies to improve the application of generative transformer models to \textsc{Pythia8} simulations of electron-proton scattering at the Electron-Ion Collider. Specifically, we propose a feature-based structured tokenization approach that utilizes multiple tokens per particle, improving expressivity, while reducing the total number of unique tokens needed. We evaluate this method against grid-based binning, K-means clustering, and vector-quantized variational auto-encoders on the event simulations. Our results show that feature-based structured tokenization reduces discretization error, leading to more accurate generative modeling of particle-level events.