Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “LLM”

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 37 records · Page 2
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

XMark: Reliable Multi-Bit Watermarking for LLM-Generated Texts

Multi-bit watermarking has emerged as a promising solution for embedding imperceptible binary messages into Large Language Model (LLM)-generated text, enabling reliable attribution and tracing of malicious usage of LLMs. Despite recent progress, existing methods still face key limitations: some become computationally infeasible for large messages, while others suffer from a poor trade-off between text quality and decoding accuracy. Moreover, the decoding accuracy of existing methods drops significantly when the number of tokens in the generated text is limited, a condition that frequently arises in practical usage. To address these challenges, we propose XMark, a novel method for encoding and decoding binary messages in LLM-generated texts. The unique design of XMark’s encoder produces a less distorted logit distribution for watermarked token generation, preserving text quality, and also enables its tailored decoder to reliably recover the encoded message with limited tokens. Extensive experiments across diverse downstream tasks show that XMark significantly improves decoding accuracy while preserving the quality of watermarked text, outperforming prior methods. The code will be made publicly available upon acceptance.

Xu, Jiahao [University of Nevada, Reno]↗
DOE OSTI2025

Verifying LLM generative agents reflect human behavior in contested information environments to effectively simulate disinformation campaigns (Proteus)

Disinformation poses a significant and evolving threat to today’s online environment. Individuals encounter challenges in detecting disinformation, subsequently influencing their behavior and decision-making processes. Our research examines the potential use of large language model (LLM) generative agents (LGAs) to replicate human behavior to better understand how disinformation is spread in online environments. Using human subjects research, we first investigate how personality traits, individual differences, and demographic factors relate to decision-making in simulated online disinformation environments. Then, we examine whether LGAs can effectively replicate human responses in the same simulated online environments when assigned personality traits, demographic characteristics and behavioral attributes. Our findings indicate that LGAs can align with human decisions in these scenarios; however, alignment is contingent upon scenario context, persona settings and LLM selection. Results provide valuable insights for methodology refinement in future research and in utilizing LGAs to model complex national security challenges such as disinformation campaigns.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Mechanical and thermomechanical properties of an LLM-105 based PBX high explosive with and without accelerated aging

In this paper we present measurements and analysis on the mechanical and thermomechanical properties of the LLM-105 based plastic bonded high explosive RX-55-DQ. Specifically we present uni-axial compression, and thermal measurements (CTE and ratchet growth), and for some of the measurements we provide a comparison to measurements on TATB-based PBX 9502. We also present the same mechanical property measurements performed on RX-55-DQ samples that underwent accelerated aging.

thermomechnical properties↗
DOE OSTI2025

Properties of different LLM-105 preparations

LLM-105 (2,6-diamino-3,5-dinitropyrazine-1-oxide) has been prepared at several different size quantities (~10–70 Kg) and not subjected to further purification, such as recrystallization. Chemical and physical properties were compared—small-scale safety testing (impact, friction, ESD), thermal (chemical reactivity, differential scanning calorimetry (DSC), thermogravimetric analysis (TGA), one-dimensional time-to-explosion (ODTX)), morphological (particle-size distribution (PSD), scanning electron microscopy (SEM), powder x-ray diffraction (PXRD), skeletal density). Each preparation had similar properties to the other preparations except for PSD and PXRD. For PSD, all preparations exhibited a mono-modal distribution with a variation in median values from ~ 40 μm to over 80 μm. All PXRD values indicated the same morphology with minor variations in crystal orientations. Possible polymorphism was also observed in a few cases. SEM images and PXRD indicate all preparations to be diamond-type and X-type crystals with some twinning. Impact, friction, and ESD and thermal sensitivities values are consistent with these type of crystal habits as seen in vast literature studies.

Explosive characterization↗
DOE OSTI2025

Rapid Adaptation of Chemical Named Entity Recognition Using Few-Shot Learning and LLM Distillation

Named entity recognition (NER) has been widely used in chemical text mining for the automatic identification and extraction of chemical entities. However, existing chemical NER systems primarily focus on scenarios with abundant training data, requiring significant human effort on annotations. This poses challenges for applications in the chemical field, such as catalysis, where many advancements have traditionally relied on trial-and-error investigations and incremental adjustment of variables. This hinders catalysis science and technology progress in addressing emerging energy and environmental crises. In this work, we propose a few-shot NER model that can quickly adapt to extract new types of chemical entities by using only a limited number of annotated examples. Our model employs a metric-learning approach to transfer entity similarity knowledge from high-resource chemical domains (with abundant annotations) to enable effective entity recognition in low-resource specialized domains (limited annotation). We validate the effectiveness of our model on a few-shot chemical NER benchmark built based on six existing chemical NER data sets. Experiments show that the proposed few-shot NER model can achieve reasonable performance with only 5 examples per entity type and shows consistent improvement as the number of examples increases. Furthermore, we demonstrate how the proposed model can be trained with large language model (LLM) annotated data, opening a new pathway for rapid adaptation of NER systems. Furthermore, our approach leverages the knowledge broadness of large language models for chemistry while distilling this knowledge into a lightweight model suitable for efficient and in-house use.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2026

The ballad of LLM agents: philosophical reasoning for chemistry

Large language models (LLMs) show remarkable potential for scientific reasoning but often produce unreliable or scientifically unactionable outputs when faced with multi-step logic, domain grounding, and interpretability challenges, especially in complex fields like chemistry and materials science. Here, we introduce a framework of philosophical reasoning agents, inspired by canonical thinkers such as Socrates, Descartes, Kant, and Hume, to guide LLM behavior via structured prompt engineering. These agents embody distinct reasoning paradigms (dialectical inquiry, deductive logic, rule-based judgment, and empirical validation) and are evaluated across multiple chemistry subdomains, physical, analytical, general, inorganic, and organic chemistry, using the ChemBench benchmark. Our agentic prompting approach yields substantial accuracy gains on open-ended numerical chemistry questions, with gains of +11.5 percentage points for GPT-4o with Hume, +4.5 percentage points for GPT-5 with Kant, and +21.8 percentage points for GPT-5.1 with Socrates at the strict 1% error threshold, relative to the corresponding base models. Beyond accuracy, we observe benchmark-level model–agent performance patterns, suggesting that different prompting styles interact differently with each base model. These findings demonstrate that embedding philosophy-of-science principles into multi-agent frameworks can improve and produce interpretable, adaptive, and domain-aligned scientific LLMs.

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

Towards Automated Reasoning Chains for Verification of LLM-Generated Scientific Code

With the rise of Large Language Model (LLM) generated code, including in domains like scientific computing, ensuring not only syntactical, but also mathematical correctness, has become a critical task. Traditional formal methods approaches often struggle with the ambiguity of floating-point code, and full symbolic execution is extremely costly and limited. We propose a chain-of-reasoning approach that iteratively lifts basic semantics from code into the SPIRAL system and then establishes numerical equivalency to the desired mathematical operation. Here, we leverage the ample mathematical knowledge already formalized in SPIRAL to enable the system to recognize not just different implementations of the same algorithm but fully separate approaches to solving the given problem. The chain establishes tight error bounds on the output of given code with respect to the true continuous solution it approximates, quantifying all sources of error. We demonstrate this approach by establishing the correctness of a pseudospectral solver for a simple 1-dimensional Poisson problem.

Oschatz, Quentin [Carnegie Mellon University,Pitts↗
DOE OSTI2026

ChatMPI: LLM-Driven MPI Code Generation for HPC Workloads

The Message Passing Interface (MPI) standard plays a crucial role in enabling scientific applications for parallel computing and is an essential component in high-performance computing (HPC). However, implementing MPI code manually—especially applying a proper domain decomposition and communication pattern—is a challenging and error-prone task. We present ChatMPI, an AI assistant for MPI parallelization of sequential C codes. In our analysis, we focus on testing six essential HPC workloads, which are based on Basic Linear Algebra Subprograms levels 1, 2, and 3 as well as sparse, stencil, and iterative operations. We analyze the process of creating ChatMPI by using the ChatHPC library. This lightweight large language model (LLM)–based infrastructure enables HPC experts to efficiently create and supervise trustworthy AI capabilities for critical HPC software tasks. We study the data required for training (fine-tuning) ChatMPI to generate parallel codes that not only use MPI syntax correctly but also apply HPC techniques to reduce memory communication and maximize performance by using proper work decomposition. With a relatively small training dataset composed of a few dozen prompts and fewer than 15 minutes of fine-tuning on one node equipped with two NVIDIA H100 GPUs, ChatMPI elevates trustworthiness for MPI code generation of current LLMs (e.g., Code Llama, ChatGPT-4o and ChatGPT 5). Additionally, we evaluate the performance of the MPI codes generated by ChatMPI in comparison with the ones generated by ChatGPT-4o and ChatGPT-5. The codes generated by ChatMPI provide up to a 4 × boost in performance by using better problem decomposition, communication patterns, and HPC techniques (e.g., communication avoiding).

Valero Lara, Pedro [ORNL] (ORCID:0000000214794310)↗
DOE OSTI2025

HALLUFIELD: DETECTING LLM HALLUCINATIONS VIA FIELD-THEORETIC MODELING

A research-focused Python package that implements our hallucination-detection method for large language models(LLMs). The code computes stability signals from LLM predictions across various hyperparameters sweeps and combines free-energy/entropy–style metrics to flag likely hallucinations, with tunable thresholds for batch scoring. The repo includes evaluation scripts, config files, and example notebooks to reproduce benchmark results and ablations; it depends on standard open-source libraries (PyTorch, Hugging Face) and runs on CPU/GPU. The repository examples only use public models/datasets only.

Bhattarai, Manish [Los Alamos National Labs]↗
DOE OSTI2025

Comparative Analysis of DNA LLM Classification Techniques Using Intra-Layer Feature Extraction with Autoencoder Stacks [Poster]

This project conducts a comparative analysis of DNA LLM classification techniques using Evo2, Grover, and UTRML, focusing on intra-layer feature extraction in Evo2. By extracting features from multiple layers of Evo2 and integrating them into an autoencoder stack with a binary classification head, we evaluate its effectiveness in classifying genomic sequences compared to smaller DNA language models. My findings demonstrate that Evo2 outperforms Grover and UTRML in classification accuracy on a dataset provided by department 08625, CAO2021, while UTRML offers competitive performance with lower computational costs. This study highlights the potential of advanced embedding techniques in enhancing genomic data analysis and informs future research in bioinformatics.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

LLM Generation of Online Courses from a Curated Set of Documents in the Nuclear Safeguards Domain

A multidisciplinary team at Argonne National Laboratory explores the application of advanced technologies to enhance knowledge transfer and retention within the nuclear safeguards domain. Specifically, it examines the feasibility of leveraging secure large language models (LLMs) to streamline the creation of e-learning modules for the U.S. National Nuclear Security Administration (NNSA) Office of International Nuclear Safeguards (NA-241). The initiative addresses the critical need for preserving institutional memory and accelerating skill development amidst the imminent retirement of senior professionals in the field in addition to supporting good knowledge management practices. The project integrates instructional design theory with cutting-edge AI technologies to transform curated document sets from the Safeguards Knowledge Repository (SKR) into modular online courses. By automating the generation of learning objectives and instructional content, the effort aims to reduce manual effort while maintaining high-quality educational outcomes. A limited measure of human supervision, however, ensures accuracy, relevance, and alignment with NNSA’s strategic priorities. Key findings highlight the potential of AI-assisted course generation to support safeguards professionals by creating structured, interactive learning experiences. The report underscores the importance of SME validation to address limitations in AI-generated content, such as terminology errors and gaps in coverage. Recommendations include adopting a structured workflow combining LLM acceleration with expert oversight to ensure accuracy, usability, and alignment with learner needs. This work demonstrates Argonne’s commitment to advancing national security and scientific excellence through innovative knowledge management solutions.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗
DOE OSTI2026

PRACtical LLM Evaluation using Performance, Response, and Context

This project produced practical methods for evaluating Large Language Models (LLMs) based only on characteristics of LLM responses and without ground truth. Further development of these methods will enable the ability to rapidly identify and mitigate different failure types and supports the appropriate use of LLMs in mission applications.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

GridCoPilot for Thermal Events: An LLM-Based Platform for Power Grid Reliability Analysis

Large Language Models show promise for translating natural language into database queries, but deploying such systems in safety-critical domains requires high reliability. We present an application of GridCoPilot to thermal event analysis (heatwaves and coldwaves) that affect power grid reliability. Our approach uses a LangChain SQL Agent to translate natural language queries into auditable SQL statements, with deterministic visualization routines that parse the structured query results. We introduce structural framing as a design principle, we integrate a NERC-region-level event library with county-level meteorology and decompose the combined data into three relational tables (event metadata, county-level event details, and a county-to-NERC subregion mapping), using prompt-guided joins to direct the model toward correct multi-table queries. For two core analytical patterns (identifying worst events by region and by region-year), the system achieved 100% SQL accuracy across all 16 NERC subregions and both event types (64 queries total). These results validate the approach for target use cases, though performance on diverse natural language formulations requires further investigation. We discuss design trade-offs, failure modes including JSON output truncation, and pathways for extending this approach to other hazard domains.

24 POWER TRANSMISSION AND DISTRIBUTION↗