Search NASA⌕ Search

SEARCH · Search NASA

Results for “scientific training”

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 199 records · Page 11
All sourcesNASA NTRSDOE OSTI
DOE OSTI2026

An information-matching approach to optimal experimental design and active learning

The efficacy of mathematical models heavily depends on the quality of the training data, yet collecting sufficient data is often expensive and challenging. Many modeling applications require inferring parameters only as a means to predict other quantities of interest (QoI). Because models often contain many unidentifiable (sloppy) parameters, QoIs often depend on a relatively small number of parameter combinations. Therefore, we introduce an information-matching criterion based on the Fisher information matrix to select the most informative training data from a candidate pool. This method ensures that the selected data contain sufficient information to learn only those parameters that are needed to constrain downstream QoIs. It is formulated as a convex optimization problem, making it scalable to large models and datasets. Here, we demonstrate the effectiveness of this approach across various modeling problems in diverse scientific fields, including power systems and underwater acoustics. Finally, we use information-matching as a query function within an active learning (AL) loop for materials science applications. In all these applications, we find that a relatively small set of optimal training data can provide the necessary information for achieving precise predictions. These results are encouraging for diverse future applications, particularly AL in large machine-learning models.

Materials science↗
DOE OSTI2025

The Institute for Nuclear Science to Inspire the next Generation of a Highly Trained workforce (INSIGHT) at FRIB

The proposed INSIGHT Center at FRIB has two objectives: (1) provide a center to support and coordinate a nationwide traineeship effort; and (2) offer traineeships at FRIB by leveraging its scientific opportunities. This will provide an environment to: (i) recruit and retain undergraduate students in (nuclear) physics and sustain and/or increase their interest, confidence, and enthusiasm in this field; (ii) provide participants with a toolset to become effective independent researchers who pursue further research opportunities as undergraduates; and (iii) encourage participants to pursue graduate studies and potential careers in nuclear science, or related STEM fields.

07 ISOTOPE AND RADIATION SOURCES↗
DOE OSTI2025

Loss Landscape Analysis for Reliable Quantized ML Models for Scientific Sensing

In this paper, we propose a method to perform empirical analysis of the loss landscape of machine learning (ML) models. The method is applied to two ML models for scientific sensing, which necessitates quantization to be deployed and are subject to noise and perturbations due to experimental conditions. Our method allows assessing the robustness of ML models to such effects as a function of quantization precision and under different regularization techniques -- two crucial concerns that remained underexplored so far. By investigating the interplay between performance, efficiency, and robustness by means of loss landscape analysis, we both established a strong correlation between gently-shaped landscapes and robustness to input and weight perturbations and observed other intriguing and non-obvious phenomena. Our method allows a systematic exploration of such trade-offs a priori, i.e., without training and testing multiple models, leading to more efficient development workflows. This work also highlights the importance of incorporating robustness into the Pareto optimization of ML models, enabling more reliable and adaptive scientific sensing systems.

Baldi, Tommaso [Pisa, Scuola Normale Superiore]↗
DOE OSTI2021

Radiological Anomaly Detection And Identification (RADAI) v1.0

The Radiological Anomaly Detection and Identification (RADAI) software package is a python library for implementing, training, and storing algorithms that detect and identify anomalies in gamma-ray spectra. The library defines a general framework for implementing detection (binary) and identification (classification) algorithms, objects to encapsulate the results of analyses, a variety of temporal filtering tools that can be used in constructing algorithms, and conceptual design that allows easy reading and writing of algorithms (and their time dependent state). In addition to this framework, the library includes implementations of a variety of algorithms from the scientific literature including: gross-counts k-sigma, SPRT, N-SCRAD, Region of Interest, and Censored Energy Window. The implementation of these algorithms within the RADAI package was done to facilitate user-initiated training and configuration to by applied to different gamma-ray detector types. Finally, benchmarked and synthetic datasets will be made available for standardized algorithm characterization with corresponding utilities in the RADAI package for data access and processing.

Joshi, Tenzing↗
DOE OSTI2023

DGaaS: GPU as a Service on Distributed Computing System

In the rapidly evolving landscape of scientific computing, Graphics Processing Units (GPUs) have become indispensable for their unparalleled ability to handle parallel tasks in complex calculations, simulations, and data analysis. Their utility is further magnified in machine learning and AI applications, where they significantly accelerate model training and predictive analytics. Within this context, the Triton Inference Server emerges as a pivotal open-source tool, specializing in AI inferencing and optimizing GPU utilization across various platforms and frameworks. This paper presents an in-depth study on distributed High Throughput Computing (HTC), specifically focusing on the HTCondor framework and its resource provisioning tools, GlideinWMS and HEPCloud. These systems enable large-scale scientific experiments like CMS and DUNE to efficiently access and utilize vast computational resources. The paper explores the core architectural components of GlideinWMS, including jobs, user pools, and worker nodes, and discusses their integration with GPUs and the Triton server. The primary aim of this research is to develop a solution that optimizes GPU utilization by leveraging Glideins and containers. This approach allows computational jobs, particularly those involving AI models, to use GPUs only when essential, thereby facilitating efficient sharing of limited GPU resources. To validate this architecture, the study conducted three key tests involving custom scripts, container-based servers, and Triton server deployments. However, the study faces challenges, notably in locating the Triton server and ensuring secure remote access. To address these issues, future work will focus on developing a proxy mechanism and enhancing security protocols. In conclusion, this study offers a comprehensive roadmap for effective and efficient GPU utilization in distributed High Throughput Computing. It aims to contribute significantly to the scientific community by solving pressing problems and implementing robust solutions in collaboration with the GlideinWMS and HEPCloud teams. The research sets the stage for a more efficient, scalable, and cost-effective paradigm in scientific computing.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Progressive transfer learning for advancing machine learning-based reduced-order modeling

Abstract To maximize knowledge transfer and improve the data requirement for data-driven machine learning (ML) modeling, a progressive transfer learning for reduced-order modeling (p-ROM) framework is proposed. A key concept of p-ROM is to selectively transfer knowledge from previously trained ML models and effectively develop a new ML model(s) for unseen tasks by optimizing information gates in hidden layers. The p-ROM framework is designed to work with any type of data-driven ROMs. For demonstration purposes, we evaluate the p-ROM with specific Barlow Twins ROMs (p-BT-ROMs) to highlight how progress learning can apply to multiple topological and physical problems with an emphasis on a small training set regime. The proposed p-BT-ROM framework has been tested using multiple examples, including transport, flow, and solid mechanics, to illustrate the importance of progressive knowledge transfer and its impact on model accuracy with reduced training samples. In both similar and different topologies, p-BT-ROM achieves improved model accuracy with much less training data. For instance, p-BT-ROM with four-parent (i.e., pre-trained models) outperforms the no-parent counterpart trained on data nine times larger. The p-ROM framework is poised to significantly enhance the capabilities of ML-based ROM approaches for scientific and engineering applications by mitigating data scarcity through progressively transferring knowledge.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

On the effectiveness of neural operators at zero-shot weather downscaling

Machine-learning (ML) methods have shown great potential for weather downscaling. These data-driven approaches provide a more efficient alternative for producing high-resolution weather datasets and forecasts compared to physics-based numerical simulations. Neural operators, which learn solution operators for a family of partial differential equations, have shown great success in scientific ML applications involving physics-driven datasets. Neural operators are grid-resolution-invariant and are often evaluated on higher grid resolutions than they are trained on, i.e., zero-shot super-resolution. Given their promising zero-shot super-resolution performance on dynamical systems emulation, we present a critical investigation of their zero-shot weather downscaling capabilities, which is when models are tasked with producing high-resolution outputs using higher upsampling factors than are seen during training. To this end, we create two realistic downscaling experiments with challenging upsampling factors (e.g., 8x and 15x) across data from different simulations: the European Centre for Medium-Range Weather Forecasts Reanalysis version 5 (ERA5) and the Wind Integration National Dataset Toolkit. While neural operator-based downscaling models perform better than interpolation and a simple convolutional baseline, we show the surprising performance of an approach that combines a powerful transformer-based model with parameter-free interpolation at zero-shot weather downscaling. We find that this Swin-Transformer-based approach mostly outperforms models with neural operator layers in terms of average error metrics, whereas an Enhanced Super-Resolution Generative Adversarial Network-based approach is better than most models in terms of capturing the physics of the ground truth data. We suggest their use in future work as strong baselines.

17 WIND ENERGY↗
DOE OSTI2025

Interpretable machine learning models classify minerals via spectroscopy

Developing methods to identify mineral species confidently and rapidly from Raman spectral analysis is critical to numerous fields. Traditionally, analysis relies on pattern matching the Raman spectrum of an unknown dataset with a supporting library of well-characterized spectral data, which may prove difficult for environmental samples that are poorly crystalline or phase mixtures. Here, we developed interpretable machine learning models that can classify uranium minerals by secondary oxyanion chemistry and other physicochemical properties based solely on Raman spectra. This new ML method produces a mineral profile of physical and chemical properties for an unknown sample and can rapidly classify or identify unknown minerals from Raman data, without the need for an exact pattern match in a spectral library. Training models are validated by 1. Strong correlation of high confidence model regions with published spectroscopic assignments and 2. Correct classification of a mineral not present in training data. Training data are from the Compendium of Uranium Raman and Infrared Experimental Spectra and available crystallographic information files within the open-source Smart Spectral Matching scientific framework. Physically meaningful classifier models can rapidly identify key structural and chemical information about unknown uranium minerals and the overall methodology is broadly applicable for mineral phases.

Machine learning↗
DOE OSTI2022

Hvac: Removing I/O Bottleneck for Large-Scale Deep Learning Applications

Scientific communities are increasingly adopting deep learning (DL) models in their applications to accelerate scientific discovery processes. However, with rapid growth in the computing capabilities of HPC supercomputers, large-scale DL applications have to spend a significant portion of training time performing I/O to a parallel storage system. Previous research works have investigated optimization techniques such as prefetching and caching. Unfortunately, there exist non-trivial challenges to adopting the existing solutions on HPC supercomputers for large-scale DL training applications, which include non-performance and/or failures at extreme scale, lack of portability and generality in design, complex deployment methodology, and being limited to a specific application or dataset. To address these challenges, we propose High-Velocity AI Cache (HVAC), a distributed read-cache layer that targets and fully exploits the node-local storage or near node-local storage technology. HVAC seamlessly accelerates read I/O by aggregating node-local or near node-local storage, avoiding metadata lookups and file locking while preserving portability in the application code. We deploy and evaluate HVAC on 1,024 nodes (with over 6000 NVIDIA V100 GPUS) of the Summit supercomputer. In particular, we evaluate the scalability, efficiency, accuracy, and load distribution of HVAC compared to GPFS and XFS-on-NVMe. With four different DL applications, we observe an average 25 % performance improvement atop GPFS and 9% drop against XFS-on-NVMe, which scale linearly and are considered the performance upper bound. We envision HVAC as an important caching library for upcoming HPC supercomputers such as Frontier.

Khan, Awais↗
DOE OSTI2025

Dissection of Carbon and Nitrogen Cycling in Post-Fire Soil Environments using a Genome- Informed Experimental Community (Final Technical Report)

Wildfires are a natural part of many forest ecosystems, with globally important carbon (C) storage and nutrient cycling consequences, and they are increasing in frequency and severity in Western North America. Forest fires affect soil C stocks in complex ways; some C is released into the atmosphere through combustion, while a large percentage of the C is added to the soil in the form of pyrogenic organic matter. Worldwide, it is estimated that 16% of soil organic matter is pyrogenic, while locally, this number may be as high as 80%. Understanding how wildfires affect soil organic matter cycling requires understanding how microbes respond to pyrogenic organic matter and other post-fire soil conditions. However, our understanding of microbial interactions within post-fire soil was in its infancy at the time of our proposal. Outstanding questions included: Which microbes are capable of degrading pyrogenic organic matter? What are the relevant genes and metabolites associated with this degradation? What are the key interactions among post-fire microbes? Key highlights of outcomes supported by this grant included training eleven early-career scientists and two early-career PIs, publication of twelve peer-reviewed papers, cross-lab collaborations that empowered complex scientific approaches, the development of an open-source automated gas sampler to drive novel insights in C cycling, enhanced understanding of post-fire microbial community dynamics, and novel genetic and molecular insights into microbial responses to fire.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Strictly Enforcing Invertibility and Conservation in CNN-Based Super Resolution for Scientific Datasets

Abstract Recently, deep convolutional neural networks (CNNs) have revolutionized image “super resolution” (SR), dramatically outperforming past methods for enhancing image resolution. They could be a boon for the many scientific fields that involve imaging or any regularly gridded datasets: satellite remote sensing, radar meteorology, medical imaging, numerical modeling, and so on. Unfortunately, while SR-CNNs produce visually compelling results, they do not necessarily conserve physical quantities between their low-resolution inputs and high-resolution outputs when applied to scientific datasets. Here, a method for “downsampling enforcement” in SR-CNNs is proposed. A differentiable operator is derived that, when applied as the final transfer function of a CNN, ensures the high-resolution outputs exactly reproduce the low-resolution inputs under 2D-average downsampling while improving performance of the SR schemes. The method is demonstrated across seven modern CNN-based SR schemes on several benchmark image datasets, and applications to weather radar, satellite imager, and climate model data are shown. The approach improves training time and performance while ensuring physical consistency between the super-resolved and low-resolution data. Significance Statement Recent advancements in using deep learning to increase the resolution of images have substantial potential across the many scientific fields that use images and image-like data. Most image super-resolution research has focused on the visual quality of outputs, however, and is not necessarily well suited for use with scientific data where known physics constraints may need to be enforced. Here, we introduce a method to modify existing deep neural network architectures so that they strictly conserve physical quantities in the input field when “super resolving” scientific data and find that the method can improve performance across a wide range of datasets and neural networks. Integration of known physics and adherence to established physical constraints into deep neural networks will be a critical step before their potential can be fully realized in the physical sciences.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2023

Neural message-passing for objective-based uncertainty quantification and optimal experimental design

Various real-world scientific applications involve the mathematical modeling of complex uncertain systems with numerous unknown parameters. Accurate parameter estimation is often practically infeasible in such systems, as the available training data may be insufficient and the cost of acquiring additional data may be high. In such cases, based on a Bayesian paradigm, we can design robust operators retaining the best overall performance across all possible models and design optimal experiments that can effectively reduce uncertainty to enhance the performance of such operators maximally. While objective-based uncertainty quantification (objective-UQ) based on MOCU (mean objective cost of uncertainty) provides an effective means for quantifying uncertainty in complex systems, the high computational cost of estimating MOCU has been a challenge in applying it to real-world scientific/engineering problems. In this work, we propose a novel scheme to reduce the computational cost for objective-UQ via MOCU based on a data-driven approach. We adopt a neural message-passing model for surrogate modeling, incorporating a novel axiomatic constraint loss that penalizes an increase in the estimated system uncertainty. As an illustrative example, we consider the optimal experimental design (OED) problem for uncertain Kuramoto models, where the goal is to predict the experiments that can most effectively enhance robust synchronization performance through uncertainty reduction. We show that our proposed approach can accelerate MOCU-based OED by four to five orders of magnitude, without any visible performance loss compared to the state-of-the-art. The proposed approach applies to general OED tasks, beyond the Kuramoto model.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2025

Machine learning for reparameterization of multi-scale closures

Scientific machine learning (ML) is becoming increasingly useful in learning closure models for multi-scale physics problems; however, many ML approaches require a vast array of training data and can struggle with generalization and interpretability. Here, rather than learning an entire closure operator, we adopt an existing reduced-dimension model of the microphysics and learn an optimal re-parameterization of the solver. We demonstrate two approaches for training the reduced dimension closure model (1) an a priori method that optimizes the closure parameterization and the neural network parameters separately and (2) an a posteriori method that simultaneously optimizes both. Using the simulation of biomass pyrolysis as a motivating example, we show that the a posteriori method achieves better target losses and is less dependent on training dataset size for generalizability. We then demonstrate the impact that implementing this reparameterization has at the macroscale, showing improved predictive performance with no modification to the underlying macroscale solvers.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
DOE OSTI2024

Gas Stopper Developments for Improved Purity and Intensity of Low-Energy, Rare Isotope Ion Beams (Final Technical Report)

This final technical report summarizes the work of the Michigan State University (MSU) team supported by grant # DE-SC0021423 awarded by the Office of Nuclear Physics, Department of Energy. Objectives: The successful fulfillment of the FRIB science mission hinges on ensuring the availability of fast, stopped, and reaccelerated beams consisting of rare isotopes. This project's research and development focus was dedicated to supporting the advancement and creation of a cutting-edge linear gas stopper. The primary aim is to efficiently convert the high-intensity fast beams of rare isotopes provided by FRIB into high-quality, low-energy beams. These beams are essential for conducting stopped beam experiments or for subsequent reacceleration. The overarching goal is to advance technology, aiming to increase the beam rate capability of the linear gas stopper for medium-to-heavy-mass rare isotopes by more than tenfold compared to the currently most effective gas stopper in operation, and to improve the purity of the delivered beams. Project Description: The existing technology employed in gas stopping devices designed for low-energy, rare-isotope beams presents limitations in both the purity of extracted beams and the intensities of injected beams. These limitations are incompatible with the requirements of the recently commissioned rare isotope beam facility, FRIB. Our research and development efforts, aligned with the previously outlined objectives, focused on addressing the most critical aspects for enhancing beam-rate capability and purity. Specifically, advanced particle-in-cell simulations were developed and integrated into a simulation pipeline to explore the efficacy of multi-layer RF carpets on increasing ion transport efficiency with high incoming beam rates that generate space charge fields which can limit it. We also explored the possibility of using a collision-induced-dissociation (CID) gas cell to break up molecular contaminant ions that are generated during the stopping process. A prototype CID gas cell was constructed and tested with beams from an offline ion source, validating the concept with the successful demonstration of breaking of molecular ions. The outcome of this research enabled the formulation a conceptual design for a next-generation linear gas stopping device specifically tailored for FRIB. This device is envisioned to deliver rare-isotope-ion beams at a rate of 10 8 particles per second or higher, accompanied by advancements in purity. Methods employed: This project leverages advancements in technologies initially designed for the Advanced Cryogenic Gas Stopper (ACGS), the current state-of-the-art linear gas stopper, through the use of new simulations and beam purification via collision-induced-dissociation. The methods include: 1. Development of a prototype low-energy, low-pressure CID gas stopper. This prototype features a thin, approximately 20 nm, Si 3 N 4 entrance window designed for dissociating stable and rare-isotope molecular ions. The goal is to enhance beam purification and overall efficiency. 2. Creation of Particle-in-cell (PIC) simulations to assess the advantages of multi-layer RF carpets and multi-point extraction for ion transport efficiency. These simulations rely on the 3DCylPIC package, specifically designed for studying devices of this nature. The goal is to quantify and mitigate ion transport losses due to space charge generated in the stopping process of large numbers of ions. 3. Perform ion transport simulations across an RF carpet using an 8-phase travelling wave and evaluate its performance. Compared to the 4-phase RF carpets used in ACGS, the 8-phase carpets will double the wavelength of the generated traveling wave allowing for larger maximum RF amplitudes that could result in improved ion transport efficiency for high-intensity incoming beams when large space charge fields are present. Impact: Tackling the primary challenges associated with transforming high-energy projectile fragment beams into low-energy beams—specifically, addressing efficiency and purity—holds significant promise for advancing FRIB science. This advancement will particularly impact precision mass measurements, laser spectroscopy of short-lived nuclei, and studies in astrophysics and nuclear reactions using reaccelerated beams. These domains play a crucial role in addressing key questions outlined in the 2023 NSAC long-range plan, spanning nuclear structure, nuclear astrophysics, and fundamental symmetries. Additionally, they contribute to addressing 10 out of the 17 benchmarks identified by the NRC RISAC. The development of a next-generation gas stopping device capable of delivering low-energy, rare-isotope beams at a rate of 10 8 particles per second, or more, with high purity holds the potential to unlock experiments that would otherwise be unfeasible. Furthermore, it is expected to reduce the time required for experiments at FRIB, thereby maximizing scientific output. The research and development activities performed as part of this project bolstered essential competencies at FRIB in beam physics and ion source technologies, provided valuable training opportunities for junior scientists.

43 PARTICLE ACCELERATORS↗
DOE OSTI2022

Transforming U.S. Particle Physics Education: A Snowmass 2021 Study

The pursuit of knowledge in particle physics requires constant learning. As new tools become available, new theories are developed, and physicists search for new answers with ever-evolving methods. However, it is the case that formal educational systems serve as the primary training grounds for particle physicists. Graduate school (and undergraduate school to a lesser extent) is where researchers learn most of the technical skills required for research, develop scientific problem-solving abilities, learn how to establish themselves in their field, and begin developing their career. It is unfortunate, then, that the skills gained by physicists during their formal education are often mismatched with the skills actually required for a successful career in physics. We performed a survey of the U.S. particle physics community to determine the missing elements of graduate and undergraduate education and to gauge how to bridge these gaps. In this contributed paper, part of the 2021-22 Snowmass Community Planning Exercise, we report the results of this survey. We also recommend several specific community actions to improve the quality of particle physics education; the "community" here refers to physics departments, national labs, professional societies, funding agencies, and individual physicists.

Bitter, O.↗
DOE OSTI2020

The Industry's Long-term, Yet Near-term, Issues: The Need for Radiation Protection Professionals - 20266

Since the discovery of X-rays and radioactivity, the civil use of nuclear technology has thrived for more than a half-century and has matured to be safer and more efficient. A balance between radiological and non-radiological knowledge and experience has been gained and retained through the years. Nevertheless, a slowly developing gap has emerged during the past few decades. National and international data show a declining trend in the existence of health physics programs and associated enrollment and training, which directly impact the pipeline of radiation professionals (i.e., educators, technicians, researchers, and regulators). This decline can be attributed to many factors including the lack of scientific inquisitiveness, declining funding, and the perception that existing protocols are sufficiently protective. Decommissioning and dismantling nuclear facilities is a complex and multi-phased process that generates large quantities of various levels of radioactive waste. With multiple facilities reaching their end of operational life, and others being shut down for economic reasons, the current pipeline for radiation protection experts cannot meet the future's need. Consequently, international organizations have proposed a variety of programs to transfer knowledge of radiation protection principles and practice. While many change-oriented efforts to counteract this trend aim to promote mobility, this is just a partial solution to the actual issue: long-term availability. This paper suggests multidisciplinary collaborations with modern technologies as a path forward to educate young professionals on the diversity of radiation-related applications to retain expert knowledge and foster innovation. (authors)

12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗
DOE OSTI2024

2024 NMDC Ambassador Training Materials [Slides]

The NMDC is a sustainable data discovery platform that promotes open science and shared-ownership across a broad and diverse community of researchers, funders, publishers, societies, and other collaborators. The NMDC aims to enable multi-omic microbiome research to accelerate scientific discovery. The NMDC is a Department of Energy funded program that is a collaboration between 3 National Laboratories: Lawrence Berkeley National Laboratory (LBNL), Los Alamos National Laboratory (LANL), and Pacific Northwest National Laboratory (PNNL).

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

An evaluation of GPT models for phenotype concept recognition

Clinical deep phenotyping and phenotype annotation play a critical role in both the diagnosis of patients with rare disorders as well as in building computationally-tractable knowledge in the rare disorders field. These processes rely on using ontology concepts, often from the Human Phenotype Ontology, in conjunction with a phenotype concept recognition task (supported usually by machine learning methods) to curate patient profiles or existing scientific literature. With the significant shift in the use of large language models (LLMs) for most NLP tasks, we examine the performance of the latest Generative Pre-trained Transformer (GPT) models underpinning ChatGPT as a foundation for the tasks of clinical phenotyping and phenotype annotation. The experimental setup of the study included seven prompts of various levels of specificity, two GPT models (gpt-3.5-turbo and gpt-4.0) and two established gold standard corpora for phenotype recognition, one consisting of publication abstracts and the other clinical observations. The best run, using in-context learning, achieved 0.58 document-level F1 score on publication abstracts and 0.75 document-level F1 score on clinical observations, as well as a mention-level F1 score of 0.7, which surpasses the current best in class tool. Without in-context learning, however, performance is significantly below the existing approaches. Our experiments show that gpt-4.0 surpasses the state of the art performance if the task is constrained to a subset of the target ontology where there is prior knowledge of the terms that are expected to be matched. While the results are promising, the non-deterministic nature of the outcomes, the high cost and the lack of concordance between different runs using the same prompt and input make the use of these LLMs challenging for this particular task.

59 BASIC BIOLOGICAL SCIENCES↗