SEARCH · Search NASA

Results for “RAG”

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 55 records · Page 3

All sources NASA NTRS DOE OSTI

DOE OSTI2024

Dynamic Retrieval Augmented Generation of Ontologies using Artificial Intelligence (DRAGON-AI)

Ontologies are fundamental components of informatics infrastructure in domains such as biomedical, environmental, and food sciences, representing consensus knowledge in an accurate and computable form. However, their construction and maintenance demand substantial resources and necessitate substantial collaboration between domain experts, curators, and ontology experts. We present Dynamic Retrieval Augmented Generation of Ontologies using AI (DRAGON-AI), an ontology generation method employing Large Language Models (LLMs) and Retrieval Augmented Generation (RAG). DRAGON-AI can generate textual and logical ontology components, drawing from existing knowledge in multiple ontologies and unstructured text sources.We assessed performance of DRAGON-AI on de novo term construction across ten diverse ontologies, making use of extensive manual evaluation of results. Our method has high precision for relationship generation, but has slightly lower precision than from logic-based reasoning. Our method is also able to generate definitions deemed acceptable by expert evaluators, but these scored worse than human-authored definitions. Notably, evaluators with the highest level of confidence in a domain were better able to discern flaws in AI-generated definitions. We also demonstrated the ability of DRAGON-AI to incorporate natural language instructions in the form of GitHub issues.These findings suggest DRAGON-AI's potential to substantially aid the manual ontology construction process. However, our results also underscore the importance of having expert curators and ontology editors drive the ontology generation process.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

DOE OSTI2025

eLog analysis for accelerators: status and future outlook

This work demonstrates electronic logbook (eLog) systems leveraging modern AI-driven information retrieval capabilities at the accelerator facilities of Fermilab, Jefferson Lab, Lawrence Berkeley National Laboratory (LBNL), SLAC National Accelerator Laboratory. We evaluate contemporary tools and methodologies for information retrieval with Retrieval Augmented Generation (RAGs), focusing on operational insights and integration with existing accelerator control systems. The study addresses challenges and proposes solutions for state-of-the-art eLog analysis through practical implementations, demonstrating applications and limitations. We present a framework for enhancing accelerator facility operations through improved information accessibility and knowledge management, which could potentially lead to more efficient operations.

Accelerator Physics↗

DOE OSTI2026

ORCHID: Orchestrated Retrieval-Augmented Classification of High-Risk Property with Intelligent Decision-Making

High-Risk Property (HRP) classification is critical at U.S. Department of Energy (DOE) sites, where inventories include sensitive and often dual-use equipment. Compliance must track evolving rules designated by various export control policies to make transparent and auditable decisions. Traditional expert-only workflows are time-consuming, backlog-prone, and struggle to keep pace with shifting regulatory boundaries. We propose ORCHID, a modular agentic framework for HRP classification that pairs retrieval-augmented generation (RAG) with human oversight to produce policy based outputs that can be audited. Small cooperating agents—retrieval, description refiner, classifier, validator, and feedback logger—coordinate via agent-to-agent messaging and invoke tools through the Model Context Protocol (MCP) for model-agnostic on-premise operation. The interface follows an "Item to Evidence to Decision" loop with step-by-step reasoning, on-policy citations, and append-only audit bundles (run-cards, prompts, evidence). In preliminary tests on real HRP cases, ORCHID improves accuracy and traceability over a non-agentic baseline while deferring uncertain items to Subject Matter Experts (SMEs). The demonstration shows single item submission, grounded citations, SME feedback capture, and exportable audit artifacts—illustrating a practical path to trustworthy LLM assistance in sensitive DOE compliance workflows.

Das, Sanjay [ORNL] (ORCID:0009000542591915)↗

DOE OSTI2024

Improving Reliability of Large Language Models for Nuclear Power Plant Diagnostics [Poster]

Large Language Models (LLMs) struggle out of the box when answering factually about detailed questions, especially in domains that are sparsely represented in their training data. This causes hallucinations and reduces reliability making it difficult for them to be used in practice. This work shows that using RAG techniques can improve factual accuracy and reliability, allowing for the application of LLMs in specialized areas, even when those areas that aren’t extensively covered in their initial training.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

DOE OSTI2025

eLog Analysis for Accelerators: Status and Future Outlook

Hellert, Thorsten [LBNL, ALS]↗

DOE OSTI2025

LLMs for Mfg.—On the State of Large Language Models and Applications to Manufacturing

Additive Manufacturing (AM), referred to as 3D printing, has emerged as a key pillar of Industry 4.0 enabling layer-by-layer fabrication of intricate geometries from CAD models. In parallel, Large Language Models (LLMs), deep learning models for natural language generation trained on vast text corpora, have demonstrated unprecedented capabilities in understanding and generating human-like text. The convergence of these trends opens new opportunities at the intersection of AM and AI/ML, where LLMs can assist engineers and researchers in design, manufacture planning, and knowledge discovery. Recent academic work has begun to explore LLM applications in AM and adjacent fields, such as material science, mechanical engineering, and design for additive manufacturing. This exploration ranges from intelligent process planning to domain-specific knowledge retrieval. This survey provides a comprehensive review of current developments, focusing on peer-reviewed literature contributions that apply, adapt, and advance LLMs in general and domain-specific domains. We analyze state-of-the-art (SOTA) techniques, such as fine-tuning foundational models for specific domains, retrieval-augmented generation (RAG) pipelines, knowledge graph integration, and delve into the architectures and evaluation methods employed. The goal of this survey is to inform researchers and practitioners of the current capabilities and limitations of LLMs in general and in domain-specific applications, and to outline how these models are being tailored to meet the requirements of these applications.

36 MATERIALS SCIENCE↗

DOE OSTI2025

ESAC v1: Enhanced AI-Powered Chatbot for EQ-SANS Experiment Automation Improvements and Updates

ESAC (EQ-SANS Assisting Chatbot) is an advanced AI-powered application designed to streamline the workflow of neutron scattering experiments at the Spallation Neutron Source (SNS). This report documents the significant advancements in ESAC v1, which include the integration of a combined In-Context Learning (ICL) + Retrieval-Augmented Generation (RAG) capability, a robust integrated development environment, and standalone executable distribution. These enhancements address the limitations of the original version, making ESAC v1 a transformative tool for researchers. The report also discusses the technical challenges encountered during development and their resolution, highlighting the impact of these improvements on the neutron scattering research community.

36 MATERIALS SCIENCE↗

DOE OSTI2025

Using Large Language Models to help customers monitor global threat data

Large Language Models have proven adept at answering general knowledge questions. To make these generative AI tools useful to our mission customers for monitoring global threats, the data sciences team at Sandia is utilizing retrieval augmented generation (RAG) techniques to customize these models with local data. The local data we use consists of data such as research articles and patent abstracts that we've collected over the last several years using automated pipelines.

Herzer, John Andrew [Sandia National Laboratories ↗

DOE OSTI2026

Augmenting LLM-Based Agents for Improved Performance in Pentesting and Commissioning Operational Technology in Critical Infrastructure

Artificial intelligence (AI), and more specifically large language models (LLMs) have the potential for use in penetration testing (“pentesting”) against devices, networks, and computer systems in information technology (IT). We explore the possibility of extending pentesting from IT systems to operational technology (OT) systems, which are more obscure than IT systems in their protocols and design. A challenge therefore exists when applying pretrained LLMs to OT systems as corpora are likely to underrepresent OT systems in comparison to other more prevalent systems. We evaluate augmentations of LLMs with various methods, especially retrieval augmented generation (RAG), to improve performance of the LLMs in the OT domain. In addition to pentesting, some of the testing of these OT devices may include commissioning to ensure that the newly installed devices work correctly. Our framework may also be applied in such cases.

97 MATHEMATICS AND COMPUTING↗

DOE OSTI2025

eLog analysis for accelerators: status and future outlook

Sulc, A. [LBL, Berkeley]↗

NASA NTRS1975

Variable features on Mars V - Evidence for crater streaks produced by wind erosion

High-resolution television pictures obtained by Mariner 9 are presented as evidence to show that the ragged dark streaks which appeared behind many dark craters several months after the end of the 1971 global dust storm resulted from wind erosion of a thin surface veneer consisting of dust-storm fallout. The pictures were taken over an area near the border between Mare Serpentis and Pandorae Fretum (approximately 30 deg S, 335 deg W). The high-resolution pictures are compared with low-resolution views of the same area, and it is shown that one very long dark streak is interrupted by a rille. It is concluded that this interruption proves that the dark streaks were produced by the present erosional mechanism.

Veverka, J.↗

NASA NTRS1976

Variable features on Mars. VI - An unusual crater streak in Mesogaea

An unusual prominent dark streak located in Mesogaea (near 8 deg N, 191 deg W) is described. Its appearance is unlike that of most dark streaks on Mars, many of which have ragged outlines, are variable on short time-scales, and are presumed to be erosional. The Mesogaea streak has a tapered smooth outline, and no changes within it were observed. It is suggested that this streak is depositional and that the low-albedo material originated within the associated crater itself. The source area is identified with a compact low-albedo region on the crater floor. Two possible origins for the dark material are suggested:(1) deflation from a recently exposed and relatively unconsolidated subsurface deposit, and (2) production of ash by a volcanic vent.

Veverka, J.↗

NASA NTRS1978

On the nature and visibility of crater-associated streaks on Mars

The paper considers Mariner 9 and Viking data that contradict Kuzmin's (1975) hypothesis that all crater-associated wind streaks on Mars are depositional and consist of unresolved barchan-like dunes. According to Kuzmin's hypothesis, any streak can appear either bright or dark relative to its surroundings depending on the sun's position. The spacecraft images, however, show examples of dark and light streaks visible at the same azimuth angle of the sun. Evidence that bright and dark streaks differ both in morphology and in character is considered. It is suggested that the common ragged dark streaks are probably erosion scars while most bright streaks probably represent accumulations of bright dust-storm fallout.

Veverka, J.↗

NASA NTRS1982

Survey of reader preferences concerning the format of NASA technical reports

A survey was conducted to determine the opinions of readers concerning the format (organization) of NASA technical reports and usage of technical report components. A survey questionnaire was sent to 513 LaRC engineers and scientists and 600 engineers and scientists from three (3) professional/technical societies. The response rates were 74 and 85 percent, respectively. The questionnaire included the order in which users read report components, the components reviewed or read to determine whether to read a report, report components which could be deleted, the desirability of a table of contents, the desirability of both a summary and abstract, the location of the symbols list and glossary, the integration of illustrative material, the preferred format for reference citations, column layout and right margin treatment, and person/voice. The results of the reader preference survey indicated that the conclusion was the component most often ready by survey respondents. The summary, conclusion, abstract, title page, and introduction were the components used most frequently to determine if a report would actually be read. Respondents indicated that a summary as well as an abstract should be included, that the definition of symbols and glossary of terms should be located in the front of the report, and that illustrative material should be integrated with the text rather than grouped at the end of the report. Citation by number was the preferred format for references. A one-column, ragged right margin was preferred. Third person, passive voice was the style of writing preferred by the respondents.

Pinelli, T. E.↗

NASA NTRS1983

The atmospheric lifetime experiment. I - Introduction, instrumentation, and overview

The Atmospheric Lifetime Experiment is designed to determine accurately the atmospheric concentrations of the four halocarbons CFCl3, CF2Cl2, CCl4, and CH3CCl3, and also of N2O with emphasis on measurement of their long-term trends in the atmosphere. Comparison of these concentrations and trends for the four halocarbons with estimates of their industrial emission rates then enables calculations of their global circulation rates and globally averaged atmospheric lifetimes. The experiment utilizes automated dual-column electron-capture gas chromatographs which sample the background air about 4 times daily at the following globally distributed sites: Adrigole, Ireland, Cape Meares, Oregon; Ragged Point, Barbados; Point Matatula, American Samoa, and Cape Grim, Tasmania. The climatology of these 'clean air' sites and their ability to describe the global air mass are reviewed. The instrumentation and methods for data acquisition and processing are then described. An overview of the data obtained and the trends derived during the 3-year period from July 1978 through June 1981 for each of the five species being measured is presented.

Prinn, R. G.↗

NASA NTRS1983

The atmospheric lifetime experiment. III - Lifetime methodology and application to three years of CFCL3 data

Observations of the chlorofluorocarbon CFCl3 obtained several times daily over the period July 1978 to June 1981 at Adrigole, Ireland; Ragged Point, Barbados; Point Matatula, American Samoa; and Cape Grim, Tasmania are reported. In addition, observations at Cape Meares, Oregon are given for the period January 1980 to June 1981. On January 1, 1980, the average mixing ratio of CFCl3 in the lower troposphere is esimated to have been 168 pptv, and this is calculated to have been increasing 5.7 percent annually. Assuming that the only destruction of CFCl3 occurs in the stratosphere, the lifetime, on January 1, 1980, estimated by a trend technique is 83 + 73, or -27 years; the lifetime estimated from the global inventory of CFCl3 is to + 89 or -25 years. The maximum likelihood current lifetime estimate obtained by combining the estimates from both analysis techniques is 78 years.

Cunnold, D. M.↗

NASA NTRS1983

The Atmospheric Lifetime Experiment. IV - Results for CF2Cl2 based on three years data

Observations of dichlorodifluoromethane obtained several times daily over the period July 1978 to June 1981 at Adrigole, Ireland (52 deg N, 10 deg W), Ragged Point, Barbados (13 deg N, 59 deg W), Point Matatula, American Samoa (14 deg S, 171 deg W), and Cape Grim, Tasmania (41 deg S, 145 deg E), are reported. Observations at Cape Meares, Oregon (45 deg N, 124 deg W), are also given for the period November 1980 to June 1981. On January 1, 1980, the average mixing ratio of dichlorodifluoromethane in the lower troposphere is estimated to have been 285 pptv and to have been increasing at 6.0 percent/year. The atmospheric lifetime of this compound is estimated from this data by adjusting its destruction rate in a two-dimensional model of the atmosphere so as to provide the best fit to the observations. Assuming destruction of CF2Cl2 in the stratosphere only, the lifetime estimate for January 1, 1980, by the inventory technique is 69 + 36 or - 18 years. The trend technique principally provides a lower limit to the lifetime of 81 years. The results suggest a need for further assessment of dichlorodifluoromethane release estimates, particularly those from the USSR and eastern Europe.

Cunnold, D. M.↗

NASA NTRS1984

Preferences on technical report format - Results of a survey

A survey of 513 engineers and scientists employed at the National Aeronautics and Space Administration Langley Research Center and 600 engineers and scientists from three professional/technical societies solicited the opinions of report users concerning the format of NASA technical reports. The results indicate that a summary as well as an abstract should be included, that the definitions of symbols and glossary of terms should be located in the front of the report, and that the illustrative material should be integrated with the text rather than grouped at the end of the report. Citation of references by number, one-column, ragged-right-margin layout, and third-person writing style are also preferred by a majority of the respondents.

Pinelli, T. E.↗