Search NASA⌕ Search

SEARCH · Search NASA

Results for “Large language 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.

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

At least 343 records · Page 19
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

HEPTAPOD: Orchestrating High Energy Physics Workflows Towards Autonomous Agency

Many workflows in high-energy-physics (HEP) stand to benefit from recent advances in transformer-based large language models (LLMs). While early applications of LLMs focused on text generation and code completion, modern LLMs now support orchestrated agency: the coordinated execution of complex, multi-step tasks through tool use, structured context, and iterative reasoning. We introduce the HEP Toolkit for Agentic Planning, Orchestration, and Deployment (HEPTAPOD), an orchestration framework designed to bring this emerging paradigm to HEP pipelines. The framework enables LLMs to interface with domain-specific tools, construct and manage simulation workflows, and assist in common utility and data analysis tasks through schema-validated operations and run-card-driven configuration. To demonstrate these capabilities, we consider a representative Beyond the Standard Model (BSM) Monte Carlo validation pipeline that spans model generation, event simulation, and downstream analysis within a unified, reproducible workflow. HEPTAPOD provides a structured and auditable layer between human researchers, LLMs, and computational infrastructure, establishing a foundation for transparent, human-in-the-loop systems.

Menzo, Tony [Alabama U.; Fermilab] (ORCID:00000002↗
DOE OSTI2025

AI for Interpreting Nuclear Power Plant Documents for Power Uprates

To reduce the cost and time needed for regulatory compliance, nuclear power plants (NPPs) can utilize artificial intelligence (AI) to assist in interpreting complex and voluminous documents that typically span thousands of pages. Usually, the process of interpreting a plant’s technical specifications (TSs) and associated documents is labor intensive. This study aims to understand what processes state-of-the-art large language models (LLMs) can automate and to identify the pitfalls associated with using LLMs to reduce human labor costs and time. This research uses a recent AI technology called retrieval augmented generation (RAG), which retrieves pages of information from TSs and associated documents to assist with NPP power uprates (cleared to produce more power). LLMs are integral to RAG because they create human-like responses based on the retrieved information, aiding in the interpretation and application processes. A baseline case demonstrates how LLMs can operate successfully for a power uprate application. Then five use cases show five types of potential failures: (1) RAG retrieving the incorrect information, (2) RAG misinterpreting the retrieved information, (3) RAG relying on knowledge not contained in the retrieved information, (4) RAG hallucinating, and (5) RAG refusing to answer. The results of the five use cases suggest that automating the human interpretation of TSs and associated documents with AI should be approached with caution. A subject-matter expert reviewed the AI outputs from the five use cases and concluded that an LLM can produce technical information that is needed to produce power uprate applications in certain instances.

21 - SPECIFIC NUCLEAR REACTORS AND ASSOCIATED PLAN↗
DOE OSTI2026

Resource-Adaptive Federated Text Generation with Differential Privacy

In cross-silo federated learning (FL), sensitive text datasets remain confined to local organizations due to privacy regulations, making repeated training for each downstream task both communication-intensive and privacy-demanding. A promising alternative is to generate differentially private (DP) synthetic datasets that approximate the global distribution and can be reused across tasks. However, pretrained large language models (LLMs) often fail under domain shift, and federated finetuning is hindered by computational heterogeneity: only resource-rich clients can update the model, while weaker clients are excluded, amplifying data skew and the adverse effects of DP noise. We propose a flexible participation framework that adapts to client capacities. Strong clients perform DP federated finetuning, while weak clients contribute through a lightweight DP voting mechanism that refines synthetic text. To ensure the synthetic data mirrors the global dataset, we apply control codes (e.g., labels, topics, metadata) that represent each client’s data proportions and constrain voting to semantically coherent subsets. This two-phase approach requires only a single round of communication for weak clients and integrates contributions from all participants. Experiments show that our framework improves distribution alignment and downstream robustness under DP and heterogeneity.

Wang, Jiayi [ORNL]↗
DOE OSTI2026

REFSafE: A RAG-Enabled Framework for Predictive Risk Analysis and Automated Safety Report Generation in Mission-Critical Environments

Operational safety in mission-critical environments requires AI systems that are accurate, interpretable, and resistant to hallucination. We present an agentic Retrieval-Augmented Generation (RAG) framework, REFSafe, for grounded hazard analysis and automated safety report generation. The system integrates Large Language Models (LLMs) with structured operational data, historical incident repositories, policy documents, and external authoritative sources. Through iterative agentic reasoning, the framework retrieves, verifies, and synthesizes evidence prior to generation, enforcing citation-backed outputs with explicit source attribution (documents, links, and prior events) to ensure traceability and trust. To mitigate hallucinations and unsupported claims, all risk assessments and forecasts are constrained to retrieved evidence, with confidence signals derived from retrieval relevance and source consistency. A transparent pipeline enables subject matter experts (SMEs) to validate predictions, and provide structured feedback, forming a continuous performance calibration loop. Preliminary deployment demonstrates improved reliability in hazard detection and safety/vulnerability report generation. This work advances trustworthy, evidence-grounded AI for predictive safety intelligence in mission-critical operations.

Das, Sanjay [ORNL] (ORCID:0009000542591915)↗
DOE OSTI2026

FATHOMS-RAG: A Framework for the Assessment of Thinking and Observation in Multimodal Systems that use Retrieval Augmented Generation

Retrieval-augmented generation (RAG) has emerged as a promising paradigm for improving factual accuracy in large language models (LLMs). We introduce a benchmark designed to evaluate RAG pipelines as a whole, evaluating a pipelines ability to ingest several modalities of information. We present (1) a curated dataset of 93 questions designed to evaluate a pipeline's ability to ingest textual data, tables, images, multimodal data, and cross-document multimodal data; (2) a phrase-level recall metric for correctness; (3) a nearest-neighbor embedding classifier in an attempt to classify pipeline hallucinations; (4) a comparative evaluation of 2 pipelines built with open-source retrieval mechanisms and 4 closed-source foundational models; and (5) a third-party human evaluation of the alignment of our correctness and hallucination metrics. We find that closed-source pipelines significantly outperform open-source pipelines in both the correctness and halucination metrics, with a wider performance gap in questions relying on multimodal and cross-document information. We also find after a human evaluation of our correctness and hallucination metric compared with our questions and pipeline responses, average agreement was 4.62 for correctness 4.53 for hallucination detection on a 1-5 Likert scale with 5 being strongly agree with our determination.

Hildebrand, Samuel [ORNL] (ORCID:0009000465963104)↗
DOE OSTI2026

Enhancing Operational Safety via Agentic Dialogue Hazard Identification Analysis

Operational safety in high-stakes domains such as industrial process control, autonomous, and safety-critical systems demand reliable hazard identification. While large language models (LLMs) have shown promise in automating safety analysis tasks, single-turn, monolithic inference is brittle: it lacks the self-correction, deliberation, and contextual refinement that safety engineers apply iteratively. In this paper, we introduce HAZDIAL, a framework that investigates whether structured agentic dialogue (multi-agent, multi-turn interactions) improves the quality of NLP-based hazard identification over single-pass baselines. We systematically compare two dialogue modalities: adversarial debate and constructive discussion, and propose an genetic algorithm-based agentic interaction optimization. We evaluate all configurations against a curated golden dataset using standard classification metrics (accuracy, precision, recall, F1) and a novel dialogue metrics. This work advances the intersection of dialogue systems, multi-agent reasoning, and AI safety, providing empirical evidence for dialogue-driven hazard analysis.

Das, Sanjay [ORNL] (ORCID:0009000542591915)↗
DOE OSTI2026

Reining in an Agentic Harness for High Energy Physics

Agentic systems now address tasks across theoretical, phenomenological, and experimental high energy physics (HEP), but their scientific capabilities remain difficult to reuse across different large language models, providers, and harnesses. We argue that stable parts of these workflows should be promoted into versioned scientific operations and exposed through common protocols. Existing general-purpose harnesses can then be specialized for HEP through task-specific sets of tools and skills, while community-maintained registries would make these capabilities discoverable and citable. We identify mismatches in conventions, assumptions, and domains of validity among independently developed operations as a potential obstacle to their composition, and discuss machine-readable scientific contracts as one possible solution. These design principles and evaluation guidelines provide a near-term path toward a portable and community-maintained agentic harness for HEP.

Menzo, Tony [Alabama U.; Fermilab] (ORCID:00000002↗
DOE OSTI2025

NEPATEC2.0: NEPA Text Corpus v2.0

The National Environmental Policy Act of 1969, as amended (NEPA), is a major environmental law in the United States, requiring Federal agencies to consider and document potential environmental impacts before deciding on a proposed action. Modernization of NEPA and permitting processes faces significant challenges due to the lack of standardized formats and interoperable systems for organizing and sharing NEPA-related information across agencies. Much of the information gathered during NEPA reviews is written into documents such as categorical exclusions, environmental assessments, and environmental impact statements, then filed in predominately independent agency file stores that may or may not be publicly accessible. The application of metadata and data standards, such as those recommended by the Council on Environmental Quality (CEQ), to NEPA documents offers a shared vocabulary and structure for key entities like projects, processes, and documents that can streamline information exchange and enhance collaboration across systems. In this work, we publicly release NEPATEC2.0, an expanded corpus of NEPA documents with associated metadata. NEPATEC2.0 encompasses approximately 120,000 documents from 60,000 projects prepared by more than 60 different agencies. Modeled to align with CEQ metadata standards, NEPATEC2.0 promotes consistency in environmental reviews and supports the ongoing effort to modernize permitting technologies by facilitating more transparent, efficient, and data-driven decision-making. Importantly, NEPATEC2.0 demonstrates the possibilities and limitations of large language model-based prompting to extract information from NEPA documents at scale.

environmental review↗
DOE OSTI2026

An agentic artificially intelligent X-ray scientist

Executing experimental tasks in both normal research laboratories and large-scale scientific facilities often requires extensive human supervision and remains a key challenge on the path to fully autonomous, artificial intelligence (AI)-driven science. Here we demonstrate a large language model-driven agent that autonomously performs X-ray sample alignment on a synchrotron beamline by planning actions, executing instrumental commands, interpreting observations and iterating towards experimental goals. Based on existing large language models with structured tool-use via the model context protocol, our AI X-ray scientist was guided and tested using an in-house-built virtual experimental setup that mirrors a six-circle diffractometer at an operational synchrotron beamline. The agentic workflow developed in the virtual environment was directly deployed on a real beamline, where it correctly identified reference reflections and determined the orientation matrix, an essential first step in any type of single-crystal scattering experiment. Our AI X-ray scientist responded effectively to unexpected experimental conditions, demonstrating adaptive problem-solving and readiness for addressing practical experimental situations. Our study provides a step towards autonomous operation across diverse experimental environments at large-scale scattering facilities.

Chen, Zhantao (ORCID:0000000319543868)↗
DOE OSTI2025

NEPATEC v2.0: Standardized Metadata and Text Corpus of National Environmental Policy Act Documents

The National Environmental Policy Act of 1969, as amended (NEPA), is a major environmental law in the United States, requiring Federal agencies to consider and document potential environmental impacts before deciding on a proposed action. Modernization of NEPA and permitting processes faces significant challenges due to the lack of standardized formats and interoperable systems for organizing and sharing NEPA-related information across agencies. Much of the information gathered during NEPA reviews is written into documents such as categorical exclusions, environmental assessments, and environmental impact statements, then filed in predominately independent agency file stores that may or may not be publicly accessible. The application of metadata and data standards, such as those recommended by the Council on Environmental Quality (CEQ), to NEPA documents offers a shared vocabulary and structure for key entities like projects, processes, and documents that can streamline information exchange and enhance collaboration across systems. In this work, we publicly release NEPATEC2.0, an expanded corpus of NEPA documents with associated metadata. NEPATEC2.0 encompasses approximately 120,000 documents from 60,000 projects prepared by more than 60 different agencies. Modeled to align with CEQ metadata standards, NEPATEC2.0 promotes consistency in environmental reviews and supports the ongoing effort to modernize permitting technologies by facilitating more transparent, efficient, and data-driven decision-making. Importantly, NEPATEC2.0 demonstrates the possibilities and limitations of large language model-based prompting to extract information from NEPA documents at scale.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

MetaText: Compositional Generalization in Deep Language Models

Compositional generalization, the ability to infer and reason using novel combinations of previously encountered entities and structures, is a trait of great utility across a variety of deep learning tasks; it implies a certain ability to reason using consistent and therefore interpretable rules, to understand the constituent parts of its input at multiple levels of granularity, to be robust to spurious differences between semantically equivalent inputs, and more. We seek to understand the degree to which existing natural language models achieve or fall short of compositional generalization across a variety of tasks, whether there are distinct types of compositional generalization in practice, and to identify avenues of intervention through which we can improve models’ compositional generalization ability. After exploring these questions using an array of models, tasks, and potential interventions we see that large, pretrained language models have encountered sufficient training data to account for a variety of fine-grained compositional behavior but still struggle to reason at the level of phrases or larger language structures. Non-intrusive, data-based interventions in the form of augmenting individual sequences with compressed versions of themselves or deriving new examples from induced grammars prove insufficient to encourage greater levels of compositional reasoning, indicating that future work might benefit most from focusing on changes to a model’s inductive bias at the architectural or loss level, or by integrating compositionality-boosting data interventions into the large-scale pretraining process itself.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

An Introduction to Word Embeddings and Language Models

Language models have advanced at a phenomenal pace over the past decade. This document provides a short introduction to terminology, word embeddings (aka low-dimensional representations), and popular large-scale language models (LMs). Word embeddings are used to represent words as numerical vectors and are context-independent, meaning a word can only have a single representation (e.g., club can only be club sandwich, not golf club ). Language models can determine the probability of a given sequence of words occurring in a sentence and can provide context to distinguish between words and phrases that sound similar. LMs are context-dependent (e.g., club can be club sandwich or golf club ) and largely fall in two main classes – autoregressive and autoencoding models. Autoregressive models are pretrained on the classic language modeling task: guess the next token having read all the previous ones. Those models can be fine-tuned and achieve great results on many tasks, the most natural application is text generation. A typical example of such models is GPT, but others include GPT-2, GPT-3, CTLR, TRANSFORMER-XL, REFORMER, XLNET. Autoencoding models are pretrained by corrupting the input tokens in some way and trying to reconstruct the original sentence. They can be fine-tuned and achieve great results on many tasks such as text generation, but their most natural application is sentence classification or token classification. A typical example of such models is BERT, but others include ROBERTA, ALBERT, XML, XML-ROBERTA, FLAUBERT AND LONGFORMER.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

MCP-enabled agentic AI workflow for building energy modelling: framework and use cases

Traditional building energy modelling workflows remain labor-intensive and error-prone, requiring specialized expertise that limits broader adoption. This paper introduces a novel Model Context Protocol (MCP)-enabled framework that connects AI assistants to EnergyPlus through MCP, a standardized interface for tool invocation and context management. Two complementary integration paradigms are presented and compared: conversational integration, where users interact through natural language while an AI assistant orchestrates MCP tools on demand, and agentic workflow integration, where specialized agents coordinate autonomously to complete multi-step tasks. Using an experimental testbed for residential buildings, the end-to-end workflows are demonstrated. The conversational approach reduced typical inspection and modification tasks from 1-2 h to under 15 min, while maintaining full transparency through visible tool invocations. The agentic approach automated parametric analysis. These demonstrations establish MCP as a foundational layer for AI-assisted building energy modelling, enabling natural language interactions with simulation tools while preserving professional oversight and decision-making authority.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2025

Solving key challenges in collider physics with foundation models

Foundation models are neural networks that are capable of simultaneously solving many problems. Large language foundation models like ChatGPT have revolutionized many aspects of daily life, but their impact for science is not yet clear. In this paper, we use a new foundation model for hadronic jets to solve three key challenges in collider physics. In particular, we show how experiments can (1) save significant computing power when developing reconstruction algorithms, (2) perform a complete uncertainty quantification for high-dimensional measurements, and (3) search for new physics with model agnostic methods using low-level inputs. In each case, there are significant computational or methodological challenges with current methods that limit the science potential of deep learning algorithms. By solving each problem, we take jet foundation models beyond proof-of-principle studies and into the toolkit of practitioners.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2025

The role of chromatin state in intron retention: A case study in leveraging large scale deep learning models

Complex deep learning models trained on very large datasets have become key enabling tools for current research in natural language processing and computer vision. By providing pre-trained models that can be fine-tuned for specific applications, they enable researchers to create accurate models with minimal effort and computational resources. Large scale genomics deep learning models come in two flavors: the first are large language models of DNA sequences trained in a self-supervised fashion, similar to the corresponding natural language models; the second are supervised learning models that leverage large scale genomics datasets from ENCODE and other sources. We argue that these models are the equivalent of foundation models in natural language processing in their utility, as they encode within them chromatin state in its different aspects, providing useful representations that allow quick deployment of accurate models of gene regulation. We demonstrate this premise by leveraging the recently created Sei model to develop simple, interpretable models of intron retention, and demonstrate their advantage over models based on the DNA language model DNABERT-2. Our work also demonstrates the impact of chromatin state on the regulation of intron retention. Using representations learned by Sei, our model is able to discover the involvement of transcription factors and chromatin marks in regulating intron retention, providing better accuracy than a recently published custom model developed for this purpose.

Biochemistry & Molecular Biology↗
DOE OSTI2024

Understanding Event Trajectories Across Massive Temporal Datasets with Word Embeddings and Visualization

In collaboration with researchers from Virginia Tech, Savannah River National Laboratory has continued development of a natural language processing pipeline to identify and extract events of interest from massive open data sources in the domain of worldwide state-sponsored civil nuclear energy. The foundation of the pipeline is built on compass aligned temporal word embedding models, whereby contextual shifts are automatically identified by comparing keyword embedding vectors across successive time windows. Within the approach, a contextual shift indicates the occurrence of a potential event of interest. However, in such a broad topical domain that captures events at a global scale, across various life cycle stages, and across numerous different technology types, a user that is monitoring events may have broad interests in capturing many different event types with varying degrees of signal. As such, the quantity of information that may be returned from an automated event extraction pipeline can be substantial, requiring manual effort to sift through the information to identify any relevant bits of information. Therefore, a more streamlined workflow that aids in directing a user toward specific information at different points in time is necessary. The workflow presented here has been developed with this concept in mind, built on top of the initial prototype event extraction pipeline, whereby a user can analyze temporal text-based data sources at multiple different contextual levels to isolate key points in time and key subdomains captured within a data corpus. Using multiple corpuses that consist of approximately 7 million Tweets and 7 million news articles, the team has extended compass aligned temporal word embedding models to establish an interconnected and hierarchical structure that relates known key words of interest to documents, local topics (i.e., within a time window), and global topics across the corpuses. All of this information is packaged into a visual analytics system that is linked to the information extraction pipeline and enables a user to identify contextual information that describes the evolution of a high dimensional embedding space across time to isolate changes of interest and explore associated events. This report demonstrates the use of these analytics and a means to fuse information across multiple datasets.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Generalist multimodal AI: A review of architectures, challenges and opportunities

Multimodal models are expected to be a critical component to future advances in artificial intelligence. Here, this field is starting to grow rapidly with a surge of new design elements motivated by the success of foundation models in natural language processing (NLP) and vision. It is widely hoped that further extending the foundation models to multiple modalities (e.g., text, image, video, sensor, time series, graph, etc.) will ultimately lead to generalist multimodal models, i.e. one model across different data modalities and tasks. However, there is little research that systematically analyzes recent multimodal models (particularly the ones that work beyond text and vision) with respect to the underling architecture proposed. Therefore, this work provides a fresh perspective on generalist multimodal models (GMMs) via a novel architecture and training configuration specific taxonomy. This includes factors such as Unifiability, Modularity, and Adaptability that are pertinent and essential to the wide adoption and application of GMMs. The review further highlights key challenges and prospects for the field and guide the researchers into the new advancements.

Artificial intelligence (AI)↗
DOE OSTI2026

From natural language to control signals: a conceptual framework for semantic channel finding in complex experimental infrastructure

Modern experimental platforms such as particle accelerators, fusion devices, telescopes, and industrial process control systems expose tens to hundreds of thousands of control and diagnostic channels, accumulated over decades of hardware evolution. Operators and AI systems alike depend on informal expert knowledge, inconsistent naming conventions, and scattered documentation to locate the signals required for monitoring, troubleshooting, and automated control, creating a persistent bottleneck for reliability, scalability, and emerging language-model-driven interfaces. We formalize semantic channel finding, the task of mapping natural-language intent to concrete control-system signals, as a general problem in complex experimental infrastructure, and introduce a four-paradigm conceptual framework to guide architecture selection based on facility-specific data regimes. The paradigms span (i) direct in-context lookup over small, curated channel dictionaries, (ii) constrained hierarchical navigation through structured trees, (iii) interactive agent exploration using iterative reasoning and tool-based database queries, and (iv) ontology-grounded semantic search that decouples channel meaning from facility-specific naming conventions. We demonstrate the practical feasibility of each paradigm through proof-of-concept implementations at four operational facilities spanning two orders of magnitude in scale: from compact free-electron lasers to large synchrotron light sources, operating under diverse control-system architectures ranging from clean hierarchical naming schemes to legacy environments with decades of heterogeneous conventions. Where evaluated against expert-curated operational queries, these instantiations achieve 90%–97% accuracy, validating the framework’s applicability across real-world deployment scenarios. To accelerate adoption across the broader scientific and industrial control-system community, we release open-source, plug-and-play implementations of all three interactive paradigms-direct lookup, hierarchical navigation, and middle-layer exploration-within the Osprey framework, together with tools for channel database generation, interactive testing, and minimal-configuration deployment. This work establishes semantic channel finding as a foundational capability for human-centric and agentic AI interfaces at large-scale facilities, providing both a systematic framework for architecture design and practical resources to enable adoption without building custom infrastructure from scratch.

channel finding↗