Search NASA⌕ Search

SEARCH · Search NASA

Results for “Language 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.

At least 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

Data Driven Approach to Public Opinion Mining on Autonomous Vehicles: Sentiment Analysis of Social Media Comments Using Large Language Models

In the realm of online identity, social media has emerged as a rich and dynamic source of user-generated content, making it an invaluable resource for understanding public sentiment on a wide range of topics. Individuals often share their raw emotions and candid opinions on these platforms without fear of judgment or backlash. In this study, we conduct a sentiment analysis on user comments collected from various online platforms, with a specific focus on discussions surrounding autonomous vehicles. Leveraging the capabilities of large language models (LLMs), we classify each comment into one of five sentiment categories: Very Negative, Negative, Neutral, Positive, and Very Positive. Our approach demonstrates the effectiveness of LLMs in capturing nuanced contextual sentiment, offering a scalable and state-of-the-art alternative to traditional manual annotation methods. The results reveal key trends and insights into public perception, enabling a deeper understanding of how autonomous vehicle technologies are received by the online community. Our findings underscore the dynamic nature of public sentiment, which is shaped not only by advances in autonomous vehicle technology but also by contextual events such as regulatory developments, political adjustment and safety incidents.

24 POWER TRANSMISSION AND DISTRIBUTION↗
DOE OSTI2026

Leveraging large language models to address data scarcity in machine learning for graphene synthesis

Machine learning in experimental materials science faces significant challenges due to the scarcity of data, which are costly and time-consuming to generate, particularly when relying on in-house experiments. Literature data mining offers a potential solution but introduces issues like mixed data quality, inconsistent formats, and non-uniform reporting of synthesis parameters, resulting in partially missing and heterogeneous features across the dataset. Here, we propose data imputation and feature engineering methods that employ pre-trained large language models (LLMs) to enhance machine learning performance on scarce, heterogeneous datasets, demonstrated on graphene CVD synthesis data and the ML-HydPARK hydrogen storage dataset. GPT models perform data imputation via tailored prompting and semantic normalization of inconsistently reported features through embeddings, for example, to harmonize the complex nomenclature of CVD substrates. Beyond yielding more diverse and richer feature representations than traditional methods such as K-nearest neighbors (KNN) and Multivariate Imputation by Chained Equations (MICE), LLM-based data imputation is evaluated against dataset characteristics and prompting strategies. We vary the level of autonomy granted to the LLM, from generic prompting that leverages pre-trained knowledge for autonomous data generation to data-informed prompting that constrains outputs using target-specific information, and demonstrate which level of autonomy yields superior imputation performance across datasets and feature types. The proposed data engineering methods markedly improve downstream performance; for example, in graphene layer number classification using a support vector machine (SVM), binary accuracy increases from 39% to 65% and ternary accuracy from 52% to 72%. Fine-tuning experiments on both datasets show that combining our proposed LLM-based data imputation and feature encoding methods with numerical machine learning predictors outperforms standalone fine-tuned LLM predictors in data-scarce settings. The proposed strategies emphasize data enhancement techniques rather than refining learning architectures or regularizing loss functions, offering a broadly applicable framework for improving machine learning performance on scarce, inhomogeneous datasets.

Chemical vapor deposition↗
DOE OSTI2025

Can Large Language Models Understand Intermediate Representations?

Intermediate Representations (IRs) are essential in compiler design and program analysis, yet their comprehension by Large Language Models (LLMs) remains underexplored. This paper presents a pioneering empirical study to investigate the capabilities of LLMs, including GPT-4, GPT-3, Gemma 2, LLaMA 3.1, and Code Llama, in understanding IRs. We analyze their performance across four tasks: Control Flow Graph (CFG) reconstruction, decompilation, code summarization, and execution reasoning. Our results indicate that while LLMs demonstrate competence in parsing IR syntax and recognizing high-level structures, they struggle with control flow reasoning, execution semantics, and loop handling. Specifically, they often misinterpret branching instructions, omit critical IR operations, and rely on heuristic-based reasoning, leading to errors in CFG reconstruction, IR decompilation, and execution reasoning. The study underscores the necessity for IR-specific enhancements in LLMs, recommending fine-tuning on structured IR datasets and integration of explicit control flow models to augment their comprehension and handling of IR-related tasks.

Jiang, Hailong↗
DOE OSTI2025

RLMolLM: Reinforcement Learning-Enhanced Language Model Framework for Inverse Molecular Design

Inverse molecular design faces significant challenges due to vast chemical space and complex property requirements. While language models show promise for molecular generation, they struggle with validity, multi-property optimization, and structural constraints. This work presents RLMolLM, a reinforcement learning framework combining Proximal Policy Optimization (PPO) with genetic algorithms to address these limitations. Our approach optimizes multiple user-specified properties including quantitative estimates of drug-likeness (QED), synthetic accessibility (SA), and ADMET (absorption, distribution, metabolism, excretion, and toxicity) endpoints without requiring complete model retraining, while maintaining capability for scaffold-constrained generation where specific substructures must be preserved. We outperform state-of-the-art methods for molecular optimization, achieving best QED scores across GDB13, Moses, and Zinc datasets with up to 31% improvement over previous methods while maintaining excellent validity, uniqueness, and novelty metrics. For simultaneous multi-property optimization, our framework achieves substantial improvements in ADMET properties including 4.5-fold reduction in hERG toxicity and enhanced Caco-2 permeability compared to Moses dataset. Under structural constraints, the framework significantly improves molecular validity while preserving scaffolds and effectively optimizing properties. In conclusion, this versatile solution advances pharmaceutical and materials molecular design through effective integration of reinforcement learning and genetic algorithms with multi-property optimization and scaffold preservation.

Genetic algorithms↗
DOE OSTI2024

Towards Next-Generation Urban Decision Support Systems through AI-Powered Construction of Scientific Ontology Using Large Language Models—A Case in Optimizing Intermodal Freight Transportation

The incorporation of Artificial Intelligence (AI) models into various optimization systems is on the rise. However, addressing complex urban and environmental management challenges often demands deep expertise in domain science and informatics. This expertise is essential for deriving data and simulation-driven insights that support informed decision-making. In this context, we investigate the potential of leveraging the pre-trained Large Language Models (LLMs) to create knowledge representations for supporting operations research. By adopting ChatGPT-4 API as the reasoning core, we outline an applied workflow that encompasses natural language processing, Methontology-based prompt tuning, and Generative Pre-trained Transformer (GPT), to automate the construction of scenario-based ontologies using existing research articles and technical manuals of urban datasets and simulations. From these ontologies, knowledge graphs can be derived using widely adopted formats and protocols, guiding various tasks towards data-informed decision support. The performance of our methodology is evaluated through a comparative analysis that contrasts our AI-generated ontology with the widely recognized pizza ontology, commonly used in tutorials for popular ontology software. We conclude with a real-world case study on optimizing the complex system of multi-modal freight transportation. Our approach advances urban decision support systems by enhancing data and metadata modeling, improving data integration and simulation coupling, and guiding the development of decision support strategies and essential software components.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2024

Large Language Models for the Creation and Use of Semantic Ontologies in Buildings: Requirements and Challenges

Semantic ontologies offer a formalized, machine-readable framework for representing knowledge, enabling the structured description of complex systems. In the building domain, the adoption of ontologies like the Brick schema has transformed how buildings and their systems are modeled by providing a standardized, interoperable language. However, the complexity and the steep learning curve involved in developing and querying semantic models present substantial challenges, often requiring a workforce with specialized expertise. This paper builds on our experience in investigating how Large Language Models (LLMs) can help address these challenges, focusing on their role in constructing and querying of semantic models, particularly using the Brick Schema. Our study outlines the requirements and metrics for evaluating the scalability and effectiveness of LLM-based tools, while also discussing the current challenges and limitations in developing such tools. Ultimately, this paper aims to orient research efforts as various groups experiment with diverse techniques, while enabling more effective comparison of emerging solutions and fostering collaboration across the field.

Mulayim, Ozan Baris↗
DOE OSTI2025

Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models

Abstract We evaluate the ability of machine learning to predict whether a hypothetical crystal structure can be synthesized and explain those predictions to scientists. Fine‐tuned large language models (LLMs) trained on a human‐readable text description of the target crystal structure perform comparably to previous bespoke convolutional graph neural network methods, but better prediction quality can be achieved by training a positive‐unlabeled learning model on a text‐embedding representation of the structure. An LLM‐based workflow can then be used to generate human‐readable explanations for the types of factors governing synthesizability, extract the underlying physical rules, and assess the veracity of those rules. These explanations can guide chemists in modifying or optimizing non‐synthesizable hypothetical structures to make them more feasible for materials design.

Kim, Seongmin [Department of Chemical and Biologic↗
DOE OSTI2025

Explainable Synthesizability Prediction of Inorganic Crystal Polymorphs Using Large Language Models

Abstract We evaluate the ability of machine learning to predict whether a hypothetical crystal structure can be synthesized and explain those predictions to scientists. Fine‐tuned large language models (LLMs) trained on a human‐readable text description of the target crystal structure perform comparably to previous bespoke convolutional graph neural network methods, but better prediction quality can be achieved by training a positive‐unlabeled learning model on a text‐embedding representation of the structure. An LLM‐based workflow can then be used to generate human‐readable explanations for the types of factors governing synthesizability, extract the underlying physical rules, and assess the veracity of those rules. These explanations can guide chemists in modifying or optimizing non‐synthesizable hypothetical structures to make them more feasible for materials design.

Kim, Seongmin [Department of Chemical and Biologic↗
DOE OSTI2025

Benchmarking large language models for materials synthesis: The case of atomic layer deposition

In this work, we introduce an open-ended question benchmark, ALDbench, to evaluate the performance of large language models (LLMs) in materials synthesis, and, in particular, in the field of atomic layer deposition, a thin film growth technique used in energy applications and microelectronics. Our benchmark comprises questions with a level of difficulty ranging from the graduate level to domain expert current with the state of the art in the field. Human experts reviewed the questions along the criteria of difficulty and specificity, and the model responses along four different criteria: overall quality, specificity, relevance, and accuracy. We ran this benchmark on an instance of OpenAI’s GPT-4o. The responses from the model received a composite quality score of 3.7 on a 1–5 scale, consistent with a passing grade. However, 36% of the questions received at least one below average score. An in-depth analysis of the responses identified at least five instances of suspected hallucination. Finally, we observed statistically significant correlations between the difficulty of the question and the quality of the response, the difficulty of the question and the relevance of the response, the specificity of the question, and the accuracy of the response as graded by the human experts. Furthermore, this emphasizes the need to evaluate LLMs across multiple criteria beyond difficulty or accuracy.

Artificial intelligence↗
DOE OSTI2025

Automating and Evaluating Large Language Models for Accurate Text Summarization Under Zero-Shot Conditions

Automated text summarization (ATS) is crucial for collecting specialized, domain-specific information. Zero-shot learning (ZSL) allows large language models (LLMs) to respond to prompts on information not included in their training, playing a vital role in this process. This study evaluates LLMs' effectiveness in generating accurate summaries under ZSL conditions and explores using retrieval augmented generation (RAG) and prompt engineering to enhance factual accuracy and understanding. We combined LLMs with summarization modeling, prompt engineering, and RAG, evaluating the summaries using the METEOR metric and keyword frequencies through word clouds. Results indicate that LLMs are generally well-suited for ATS tasks, demonstrating an ability to handle specialized information under ZSL conditions with RAG. However, web scraping limitations hinder a single generalized retrieval mechanism. While LLMs show promise for ATS under ZSL conditions with RAG, challenges like goal misgeneralization and web scraping limitations need addressing. Future research should focus on solutions to these issues.

Priebe Mendes Rocha, Maria Eduarda [ORNL]↗
DOE OSTI2026

A multimodal large language model for materials science

Understanding and predicting the properties of inorganic materials is crucial for accelerating advancements in materials science and driving applications in energy, electronics and beyond. Integrating material structure data with language-based information through multimodal large language models (LLMs) offers great potential to support these efforts by enhancing human–artificial intelligence interaction. However, a key challenge lies in integrating atomic structures at full resolution into LLMs. In this work, we introduce MatterChat, a versatile structure-aware multimodal LLM that unifies material structural data and textual inputs into a single cohesive model. MatterChat uses a bridging module to effectively align a pretrained universal machine learning interatomic potential with a pretrained LLM, reducing training costs and enhancing flexibility. Our results demonstrate that MatterChat greatly improves performance in material property prediction and human–artificial intelligence interaction, surpassing general-purpose LLMs such as GPT-4. We also demonstrate its usefulness in applications such as more advanced scientific reasoning and step-by-step material synthesis.

Tang, Yingheng [Lawrence Berkeley National Laborat↗
DOE OSTI2026

Discovery of hydrogen storage molecules using large language models and machine learning

Accelerating the discovery of new molecules with targeted properties is a central challenge in molecular design. In this contribution, we present an AI-driven molecular discovery framework that integrates Large Language Models (LLMs) for generative molecular design with Machine Learning (ML)-based screening to identify novel Liquid Organic Hydrogen Carrier (LOHC) candidates. Using the developed framework, LOHC molecules were systematically generated, evaluated, and refined iteratively, combining LLM-guided molecular generation and ML-predicted hydrogenation enthalpies (Δ H ), under physicochemical property constraints such as optimal melting points (MP), desired hydrogen storage capacity (wt% H 2 ), and synthetic accessibility (SA) scores. This approach enabled the discovery of 42 new LOHC candidates in two distinct campaigns, one seeded with experimentally known and another with previously computationally identified LOHCs, respectively. Although we began with different numbers of starting molecules (31 vs . 7 seed molecules), both runs yielded a comparable number of viable candidates, suggesting an influence of chemically intuitive seed molecule selection for success. Selected LOHC molecules, such as 3-methyl pyridine, 1-ethylnapthalene, 1,1-diphenylethane, and benzofuran, were experimentally tested and compared with benchmark LOHCs (toluene and 9-ethylcarbazole) for hydrogenation using a series of commercial supported metal catalysts. The order of conversion into fully hydrogenated products at 200 °C was 3-methyl pyridine (100%) > 9-ethyl carbazole (86.4%) > 2,3-benzofuran (74%) > 1,1-diphenylethane (66.9%) > 1-ethylnapthalene (66.7%) > toluene (57%), further validating the AI-guided molecular design. This study demonstrates promise of LLM-driven molecular design in conjunction with ML-based screening for accelerated discovery and design of molecules.

Harb, Hassan [Argonne National Laboratory (ANL), A↗
DOE OSTI2025

Large Language Model Integration for Knowledge Retrieval and Interaction for the DUNE Experiment

The Deep Underground Neutrino Experiment (DUNE) is a next-generation neutrino experiment that will generate an unprecedented volume of heterogeneous information-from documentation and technical notes to experimental data and reconstruction pipelines. Efficient knowledge retrieval and contextual understanding are increasingly critical for collaboration-wide productivity and onboarding. In this work, we present DUNE-GPT, a prototype framework that leverages large language models (LLMs) and retrieval-augmented generation (RAG) to enable natural-language querying of DUNE's internal documentation and technical resources. The system provides an intelligent interface for DUNE collaborators to interact with experiment-specific knowledge while maintaining data privacy and infrastructure compliance within Fermilab computing resources.

Rafique, A. [Argonne (main)]↗
DOE OSTI2026

QCalEval: Benchmarking Vision-Language Models for Quantum Calibration Plot Understanding

Quantum computing calibration depends on interpreting experimental data, and calibration plots provide the most universal human-readable representation for this task, yet no systematic evaluation exists of how well vision-language models (VLMs) interpret them. We introduce QCalEval, the first VLM benchmark for quantum calibration plots: 243 samples across 87 scenario types from 22 experiment families, spanning superconducting qubits and neutral atoms, evaluated on six question types in both zero-shot and in-context learning settings. The best general-purpose zero-shot model reaches a mean score of 72.3, and many open-weight models degrade under multi-image in-context learning, whereas frontier closed models improve substantially. A supervised fine-tuning ablation at the 9-billion-parameter scale shows that SFT improves zero-shot performance but cannot close the multimodal in-context learning gap. As a reference case study, we release NVIDIA Ising Calibration 1, an open-weight model based on Qwen3.5-35B-A3B that reaches 74.7 zero-shot average score.

Cao, Shuxiang↗
DOE OSTI2025

Decoding substance use disorder severity from clinical notes using a large language model

Substance use disorder (SUD) poses a major concern due to its detrimental effects on health and society. SUD identification and treatment depend on a variety of factors such as severity, co-determinants (e.g., withdrawal symptoms), and social determinants of health. Existing diagnostic coding systems used by insurance providers, like the International Classification of Diseases (ICD-10), lack granularity for certain diagnoses, but American clinicians will add this granularity (as that found within the Diagnostic and Statistical Manual of Mental Disorders classification or DSM-5) as supplemental unstructured text in clinical notes. Traditional natural language processing (NLP) methods face limitations in accurately parsing such diverse clinical language. Large language models (LLMs) offer promise in overcoming these challenges by adapting to diverse language patterns. This study investigates the application of LLMs for extracting severity-related information for various SUD diagnoses from clinical notes. We propose a workflow employing zero-shot learning of LLMs with carefully crafted prompts and post-processing techniques. Through experimentation with Flan-T5, an open-source LLM, we demonstrate its superior recall compared to the rule-based approach. Focusing on 11 categories of SUD diagnoses, we show the effectiveness of LLMs in extracting severity information, contributing to improved risk assessment and treatment planning for SUD patients.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2026

A unified large language model–based framework for heterogeneous PV image diagnosis

With advances in imaging technologies, modern photovoltaic (PV) systems generate large volumes of heterogeneous image data, including visible, electroluminescence (EL), and infrared (IR) images. Existing PV image analysis models, particularly deep learning approaches, are typically task-specific and lack cross-modality generalization. To address this limitation, this paper proposes an open-source large language model (LLM)–based unified framework for heterogeneous PV image diagnostics. Through task-aware diagnostic prompting, the framework enables analysis of visible, EL, and IR images within a single pipeline, supporting both zero-shot and few-shot inference and binary and multiclass classification. It is compatible with state-of-the-art multimodal LLMs, including ChatGPT, Gemini, Claude, Qwen, and CLIP. The framework is evaluated on PV module condition classification (clean, soiling, snow, hail, and bird droppings) using visible images, cell crack detection using EL images, and hotspot detection using IR images. GPT-5.1 in few-shot mode achieves the best performance, with classification accuracy exceeding 97.3%. Open-source models such as Qwen and CLIP also deliver competitive results on visible images (around 90% accuracy), though their performance is more limited on EL and IR modalities. On the full ELPV dataset, the framework achieves 83.5% zero-shot accuracy, within 2.8% of the supervised CNN baseline, confirming scalability to larger benchmarks. Practical aspects such as reproducibility, response latency, and confidence estimation are systematically analyzed. The framework operates across PV image modalities without modality- or task-specific training, making it well suited as a rapid pre-screening tool to support downstream detailed diagnostics. A benchmark dataset of diverse labeled PV images is also released.

Li, Baojie↗
DOE OSTI2024

VLM4Bio: A Benchmark Dataset to Evaluate Pretrained Vision-Language Models for Trait Discovery from Biological Images

Images are increasingly becoming the currency for documenting biodiversity on the planet, providing novel opportunities for accelerating scientific discoveries in the field of organismal biology, especially with the advent of large vision-language models (VLMs). We ask if pre-trained VLMs can aid scientists in answering a range of biologically relevant questions without any additional fine-tuning. In this paper, we evaluate the effectiveness of 12 state-of-the-art (SOTA) VLMs in the field of organismal biology using a novel dataset, VLM4Bio, consisting of 469K question8 answer pairs involving 30K images from three groups of organisms: fishes, birds, and butterflies, covering five biologically relevant tasks. We also explore the effects of applying prompting techniques and tests for reasoning hallucination on the performance of VLMs, shedding new light on the capabilities of current SOTA VLMs in answering biologically relevant questions using images

Maruf, M [Virginia Tech, Blacksburg]↗
DOE OSTI2025

Toward particle accelerator machine state embeddings as a modality for large language models

Understanding and diagnosing the state of a particle accelerator requires navigating high-dimensional control system data, often involving hundreds of interdependent parameters. We propose a novel multimodal embedding framework that jointly learns representations of machine states from both numerical control system readouts and natural language descriptions. This enables the translation of complex machine conditions into human-readable summaries while maintaining fidelity to the underlying physical system. The obtained embeddings are subsequently adapted to an open-weights large language model via cross-attention conditioning. We demonstrate a first implementation trained on European XFEL machine state data. This work covers the embedding model architecture, training methodology, and presents initial examples demonstrating the model's capabilities in action. Due to the general concept of machine state, the model can be easily adapted to other facilities and control system environments.

Accelerator Physics↗