Search NASA⌕ Search

SEARCH · Search NASA

Results for “generative 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 127 records · Page 7
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Synthetic Scientific Image Generation with VAE, GAN, and Diffusion Model Architectures

Generative AI (genAI) has emerged as a powerful tool for synthesizing diverse and complex image data, offering new possibilities for scientific imaging applications. This review presents a comprehensive comparative analysis of leading generative architectures, ranging from Variational Autoencoders (VAEs) to Generative Adversarial Networks (GANs) on through to Diffusion Models, in the context of scientific image synthesis. We examine each model's foundational principles, recent architectural advancements, and practical trade-offs. Our evaluation, conducted on domain-specific datasets including microCT scans of rocks and composite fibers, as well as high-resolution images of plant roots, integrates both quantitative metrics (SSIM, LPIPS, FID, CLIPScore) and expert-driven qualitative assessments. Results show that GANs, particularly StyleGAN, produce images with high perceptual quality and structural coherence. Diffusion-based models for inpainting and image variation, such as DALL-E 2, delivered high realism and semantic alignment but generally struggled in balancing visual fidelity with scientific accuracy. Importantly, our findings reveal limitations of standard quantitative metrics in capturing scientific relevance, underscoring the need for domain-expert validation. We conclude by discussing key challenges such as model interpretability, computational cost, and verification protocols, and discuss future directions where generative AI can drive innovation in data augmentation, simulation, and hypothesis generation in scientific research.

Generative Adversarial Networks↗
DOE OSTI2026

Generative large language models for predictive maintenance planning

Maintenance planning and the generation of necessary components for tasks can prove time-consuming and complex. Automating the creation of recurring or similar tasks by leveraging previous planning packages and data, while uncovering insights to automate planning package generation, presents an opportunity to conserve valuable time and resources. This work aims to harness the textual and probabilistic capabilities of large language models (LLMs) to automate the generation of planning packages. Utilizing diverse data sources ranging from raw data to handwritten text, both singular and collaborative LLMs are trained and tested. Results demonstrate their capability to generate essential planning package components, effectively replicating the statistical patterns in the data. This demonstrates the use of these tools inside a digital asset for automated planning. This work outlines a methodology for constructing datasets, a training suite, and evaluation methods for LLM-based textual and conversational planning tools utilized in an asset digital twin. Results indicate that the fine-tuned models generate estimated planning information within the statistical ranges observed in real maintenance data. The models achieve high accuracy (>90%) in document question-answering and instruction generation tasks. Furthermore, the conversational retrieval-augmented generation (RAG) assistant system achieves 100% document retrieval accuracy, while conversational information capture exceeds 98% across the majority of work-package assistant modules.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Artificial intelligence tools for enzyme engineering and metabolic engineering

Enzyme engineering and metabolic engineering drive innovation in energy biotechnology. In recent years, artificial intelligence (AI) has supported successful applications in designing effective enzymes and productive microbial cell factories. This review summarizes recent advances in enzyme redesign using protein language models, de novo enzyme design with generative models, and AI tools for engineering metabolism and related cellular phenotypes. Across these areas, AI models are shifting from single modality inputs to integrated representations of protein function, metabolic pathways, and cell states. We emphasize that unifying the diverse data representations across scales will be necessary for advancements in energy biotechnology.

Volk, Michael [Univ. of Illinois at Urbana-Champai↗
DOE OSTI2025

Generative Vulnerability Assessment for Cyber-Physical Systems

Cyber-physical systems (CPS) are highly susceptible to malicious attacks due to their complex dynamics and interconnectivity. A comprehensive understanding of their vulnerabilities is essential for designing effective resilience measures. This paper presents a data-driven attack generative system for evaluating the vulnerability of CPS. The proposed approach formulates the vulnerability assessment problem as determining the feasibility of a specific attack set based on two boundary functions that represent the effectiveness and stealthiness of attacks. The attack generative model is trained using a custom loss function, with two universal approximators designed to learn the effectiveness and stealthiness functions simultaneously. Theoretical results for successful generation and asymptotic convergence of the resulting training algorithm are given. As a result, the proposed approach is evaluated via numerical simulation of an IEEE 14-bus system and gas pipeline systems, demonstrating its viability in learning how to attack nonlinear CPS and identify potential vulnerabilities.

Computer systems organization↗
DOE OSTI2025

From clutter to clarity: Emergent neural operators via questionnaire metrics

Real-world datasets in chemical engineering and bioengineering processes—such as those from catalytic reactors, multiphase flows, polymerization reactors, bioreactors, and clinical trials—can often be unlabeled or disorganized, rendering the training of existing supervised learning models ineffective at learning the underlying dynamics. To salvage these datasets for decision-making, we first seek to obtain clarity from the cluttered data. Here, we present a framework for developing “structural” generative models, discovering emergent equations, and constructing efficient emulators from scrambled datasets by integrating unsupervised organizational learning techniques (Questionnaires) with advanced deep learning architectures (Deep Hidden Physics Models and Deep Operator Networks). Our approach is demonstrated on two illustrative model systems: (a) a 1D advection–diffusion partial differential equation representing a winding underground pipe and (b) an ensemble of Stuart–Landau oscillators, an agent-based system of coupled ordinary differential equations. In both cases, we successfully reconstruct meaningful spatial, temporal, and parameter embeddings from scrambled data, enabling good predictions of system dynamics. As a result, we highlight the framework’s potential for broader applications, enabling data-driven system identification in fields with inherently disorganized or hidden parameter spaces.

42 ENGINEERING↗
DOE OSTI2025

Physics-Driven Construction of Compact Primitive Gaussian Density Fitting Basis Sets

We present a model-assisted density fitting (MADF) basis set generator, an algorithm for generating primitive atomic Gaussian density fitting (DF) basis sets (DFBSs) from a contracted Gaussian orbital basis set (OBS). The MADF algorithm produces DFBSs suitable for accurate robust DF approximation of 2-particle interactions in mean-field and correlated electronic structures. The algorithm is designed to (a) saturate the OBS product space by a large regularized set of primitive solid-harmonic Gaussian shells with nonuniform distribution of exponents, followed by (b) pruning of the shells according to their contributions to the 2- body energy of a correlated atomic ensemble. Building the DFBS generator model almost exclusively on mathematical and physical principles allows one to limit the number of parameters that control the density fitting error to three, with a single set of parameters sufficient for computations with all basis cardinal numbers, with and without correlation of core electrons, with and without scalar and spin-dependent relativistic effects, spanning almost all of the Periodic Table. Performance assessment included basis sets up to quadruple-ζ quality from several major basis set families, using molecules composed of main-group, d-block, and f-block elements. The resulting DF errors in Hartree−Fock and second-order MP2 energies (with relativistic all-electron treatments, when appropriate) were on the order of 20 and 10 μE h per electron, respectively.

Approximation↗
DOE OSTI2024

Evaluation of Converter Performance Considering Static and Dynamic Device Part-to-Part Variability

This paper presents a methodology to incorporate and analyze the impact of semiconductor device part-to-part variation on power converter performance. By integrating extensive static and dynamic device characterization data with an automated compact model generation process that reflects manufacturing variability, device models with inherent variability features are utilized in converter simulations for a comprehensive assessment of performance impacts. The traditional converter performance evaluation process typically yields fixed efficiency values, often dismissing the inherent part-to-part variability caused by the manufacturing process of semiconductor devices. To address this limitation, a large population of devices was characterized to capture variations in static parameters-such as transfer, output, and capacitance characteristics-as well as dynamic behaviors, including switching losses. This data-driven approach enables the development of individual compact models, which were then integrated into converter simulations to evaluate efficiency ranges rather than single point estimated values. The converter simulation results show that part-to-part component variation can lead to significant efficiency deviations, exceeding several percentage points in high-power conversion applications. By offering a more accurate representation of converter behavior under real-world manufacturing conditions, this methodology enables designers to anticipate performance variability, improving the robustness of power converter designs.

device characterization↗
DOE OSTI2025

Is tokenization needed for masked particle modeling?

In this work, we significantly enhance masked particle modeling (MPM), a self-supervised learning scheme for constructing highly expressive representations of unordered sets relevant to developing foundation models for high-energy physics. In MPM, a model is trained to recover the missing elements of a set, a learning objective that requires no labels and can be applied directly to experimental data. We achieve significant performance improvements over previous work on MPM by addressing inefficiencies in the implementation and incorporating a more powerful decoder. We compare several pre-training tasks and introduce new reconstruction methods that utilize conditional generative models without data tokenization or discretization. We show that these new methods outperform the tokenized learning objective from the original MPM on a new test bed for foundation models for jets, which includes using a wide variety of downstream tasks relevant to jet physics, such as classification, secondary vertex finding, and track identification.

conditional generative models↗
DOE OSTI2026

SimLBR: Learning to Detect Fake Images by Learning to Detect Real Images

The rapid advancement of generative models has made the detection of AI-generated images a critical challenge for both research and society. Recent works have shown that most state-of-the-art fake image detection methods overfit to their training data and catastrophically fail when evaluated on curated hard test sets with strong distribution shifts. In this work, we argue that it is more principled to learn a tight decision boundary around the real image distribution and treat the fake category as a sink class. To this end, we propose SimLBR, a simple and efficient framework for fake image detection with Latent Blending Regularization (LBR). Our method significantly improves cross-generator generalization, achieving up to +24.85% accuracy and +69.62% recall on the challenging Chameleon benchmark. SimLBR is also highly efficient, training orders of magnitude faster than existing approaches. Furthermore, we emphasize the need for reliability-oriented evaluation in fake image detection, introducing risk-adjusted metrics and worst-case estimates to better assess model robustness. All the code and models are availabe at: https://github.com/mvrl/SimLBR

Dhakal, Aayush [Washington University, St. Louis]↗
DOE OSTI2024

DiffLense: a conditional diffusion model for super-resolution of gravitational lensing data

Abstract Gravitational lensing data is frequently collected at low resolution due to instrumental limitations and observing conditions. Machine learning-based super-resolution techniques offer a method to enhance the resolution of these images, enabling more precise measurements of lensing effects and a better understanding of the matter distribution in the lensing system. This enhancement can significantly improve our knowledge of the distribution of mass within the lensing galaxy and its environment, as well as the properties of the background source being lensed. Traditional super-resolution techniques typically learn a mapping function from lower-resolution to higher-resolution samples. However, these methods are often constrained by their dependence on optimizing a fixed distance function, which can result in the loss of intricate details crucial for astrophysical analysis. In this work, we introduce DiffLense , a novel super-resolution pipeline based on a conditional diffusion model specifically designed to enhance the resolution of gravitational lensing images obtained from the Hyper Suprime-Cam Subaru Strategic Program (HSC-SSP). Our approach adopts a generative model, leveraging the detailed structural information present in Hubble space telescope (HST) counterparts. The diffusion model, trained to generate HST data, is conditioned on HSC data pre-processed with denoising techniques and thresholding to significantly reduce noise and background interference. This process leads to a more distinct and less overlapping conditional distribution during the model’s training phase. We demonstrate that DiffLense outperforms existing state-of-the-art single-image super-resolution techniques, particularly in retaining the fine details necessary for astrophysical analyses.

Computer Science↗
DOE OSTI2026

AI-Driven Frameworks for Characterizing Urban Energy Systems

We develop AI-driven frameworks to characterize urban energy systems with the goal of transforming planning by reducing the labor of model generation, scaling scenario exploration, and improving accuracy for localized analysis. The approach integrates top-down and bottom-up data to train different AI models that predict missing information and generate inputs and targeted scenarios for district-scale models. The result is a scalable framework that provides actionable insights for reliable and efficient planning.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗
DOE OSTI2024

The 1.3 Å resolution structure of the truncated group Ia type IV pilin from Pseudomonas aeruginosa strain P1

The type IV pilus is a diverse molecular machine capable of conferring a variety of functions and is produced by a wide range of bacterial species. The ability of the pilus to perform host-cell adherence makes it a viable target for the development of vaccines against infection by human pathogens such as Pseudomonas aeruginosa . Here, the 1.3 Å resolution crystal structure of the N-terminally truncated type IV pilin from P. aeruginosa strain P1 (ΔP1) is reported, the first structure of its phylogenetically linked group (group I) to be discussed in the literature. The structure was solved from X-ray diffraction data that were collected 20 years ago with a molecular-replacement search model generated using AlphaFold ; the effectiveness of other search models was analyzed. Examination of the high-resolution ΔP1 structure revealed a solvent network that aids in maintaining the fold of the protein. On comparing the sequence and structure of P1 with a variety of type IV pilins, it was observed that there are cases of higher structural similarities between the phylogenetic groups of P. aeruginosa than there are between the same phylogenetic group, indicating that a structural grouping of pilins may be necessary in developing antivirulence drugs and vaccines. These analyses also identified the α–β loop as the most structurally diverse domain of the pilins, which could allow it to serve a role in pilus recognition. Studies of ΔP1 in vitro polymerization demonstrate that the optimal hydrophobic catalyst for the oligomerization of the pilus from strain K122 is not conducive for pilus formation of ΔP1; a model of a three-start helical assembly using the ΔP1 structure indicates that the α–β loop and the D-loop prevent in vitro polymerization.

Bragagnolo, Nicholas↗
DOE OSTI2024

A guide on modeling electrostatics of semiconductor detectors in COMSOL Multiphysics ® for DRiFT

Adding a new semiconductor detector into the detector response function toolkit (DRiFT) requires a model of the electric potential and the electric field. To get an accurate electrostatics model, the geometry of the detector should be modeled as closely to the real geometry as feasible, including the semiconductor materials, doping layers and concentrations, and contacts. Since the electric field and potential are used to calculate the induced signal, the results of the detector response functions are largely influenced by the electrostatics models. Initial electrostatics models were completed using Silvaco, however, this document provides guidance on using COMSOL Mulitphysics ® (COMSOL) to model the detectors to give users additional flexibility. Guidance on using COMSOL and its user interface are mostly left out of this document, but best practices for geometry, modeling methods, and data format/exporting are included to help users generate models that are compatible with DRiFT, prevents bugs, issues, and inaccuracies during charge collection calculations. The following summary provides an overview of the document.

42 ENGINEERING↗
DOE OSTI2025

Contrasting Time-Frequency Representations for Unknown Waveform Detection

Identifying unseen electromagnetic waveforms is critical for many applications, like interference management, electronic warfare and spectrum management. Traditionally this is done using statistical methods for anomaly detection, which has evolved to deep learning models for identifying the unseen data, formally termed as open set recognition. Some prior methods use a generative model to emulate open set data, which face challenges in generating synthetic samples for open set while simultaneously selecting an optimal discriminator for accurate classification. To alleviate this issue, we propose a discriminative model that effectively combines time and frequency domain features of communication signals for accurate predictions. We further introduce a cosine similarity loss that makes the domain specific features unique to enhance the prediction rate. Additionally, our model avoids generic feature vectors by extracting class-specific features during training, resulting in improved class representation. The experiment results show that this combined feature approach with cosine loss outperforms single-domain models and improves accuracy by 10% over models without cosine loss.

99 - GENERAL AND MISCELLANEOUS↗
DOE OSTI2026

Hierarchical transfer learning: an agile and equitable strategy for machine-learning interatomic models

Machine-learned interatomic models are growing in popularity due to their ability to afford near quantum-accurate predictions for complex phenomena with orders-of-magnitude greater computational efficiency. However, these models struggle when applied to systems of many element types due to the approximately exponential increase in number of parameters that must be determined. To mitigate this challenge, we present a new hierarchical transfer learning approach that allows the fitting problem to be decomposed into smaller independent and reusable parameter blocks that enable development of explicitly chemically extensible ML-IAM. Application of this strategy is demonstrated for C and N mixtures under conditions ranging from nominally ambient to ~10,000 K and 200 GPa for compositions from 0 to 100% N. Ultimately, this strategy makes model generation for chemically complex systems more tractable and efficient, facilitates comprehensive model validation, and makes ML-IAM development for problems of this nature more accessible to users with limited access to extreme computing infrastructure.

Lindsey, Rebecca K. [Univ. of Michigan, Ann Arbor,↗
DOE OSTI2024

ATAT: Astronomical Transformer for time series and Tabular data

Context. The advent of next-generation survey instruments, such as theVera C. RubinObservatory and its Legacy Survey of Space and Time (LSST), is opening a window for new research in time-domain astronomy. The Extended LSST Astronomical Time-Series Classification Challenge (ELAsTiCC) was created to test the capacity of brokers to deal with a simulated LSST stream. Aims. Our aim is to develop a next-generation model for the classification of variable astronomical objects. We describe ATAT, the Astronomical Transformer for time series And Tabular data, a classification model conceived by the ALeRCE alert broker to classify light curves from next-generation alert streams. ATAT was tested in production during the first round of the ELAsTiCC campaigns. Methods. ATAT consists of two transformer models that encode light curves and features using novel time modulation and quantile feature tokenizer mechanisms, respectively. ATAT was trained on different combinations of light curves, metadata, and features calculated over the light curves. We compare ATAT against the current ALeRCE classifier, a balanced hierarchical random forest (BHRF) trained on human-engineered features derived from light curves and metadata. Results. When trained on light curves and metadata, ATAT achieves a macro F1 score of 82.9 ± 0.4 in 20 classes, outperforming the BHRF model trained on 429 features, which achieves a macro F1 score of 79.4 ± 0.1. Conclusions. The use of transformer multimodal architectures, combining light curves and tabular data, opens new possibilities for classifying alerts from a new generation of large etendue telescopes, such as theVera C. RubinObservatory, in real-world brokering scenarios.

Astronomy & Astrophysics↗
DOE OSTI2024

Feasibility of Formulating Ecosystem Biogeochemical Models From Established Physical Rules

Abstract To improve the predictive capability of ecosystem biogeochemical models (EBMs), we discuss the feasibility of formulating biogeochemical processes using physical rules that have underpinned the many successes in computational physics and chemistry. We argue that the currently popular empirically based approaches, such as multiplicative empirical response functions and the law of the minimum, will not lead to EBM formulations that can be continuously refined to incorporate improved mechanistic understanding and empirical observations of biogeochemical processes. Instead, we propose that EBM parameterizations, as a lossy data compression problem, can be better formulated using established physical rules widely used in computational physics and chemistry, and different biogeochemical processes can be more robustly integrated within a reactive‐transport framework. Through several examples, we demonstrate how mathematical representations derived from physical rules can improve understanding of relevant biogeochemical processes and enable more effective communication between modelers, observationalists, and experimentalists regarding essential questions, such as what measurements are needed to meaningfully inform models and how can models generate new process‐level hypotheses to test in empirical studies. Finally, while empirical models with more parameters are often less robust, physical rules‐based models can be more robust and show lower predictive equifinality, stemming from their enhanced consistency in representations of processes, interactions and spatial scaling.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Modeling household-level party composition behavior for multiparty activities: a random parameter nested logit modeling approach

This study presents findings of a household-level party composition model for multiparty activities. It exploits data from a comprehensive Household Travel Survey conducted by Chicago Metropolitan Agency of Planning. The study estimates a random parameter nested logit model to capture households’ unobserved preference heterogeneity and non-proportional substitution patterns in terms of activity party composition for multiparty activities. A wide variety of household demographics, activity attributes and residential neighborhood characteristics are examined in this paper. The magnitude of the impacts of the determinants are tested in this study by analyzing the elasticity of the variables, which suggests that household demographics and attributes of the multiparty activities have significant effects on the household-level activity party composition. Residential neighborhood characteristics, although somewhat less impactful, still play a meaningful role. This model will be implemented within the POLARIS transportation systems simulator to improve the activity generation modeling workflow, and the prediction accuracy of various activity-travel components.

activity party composition↗