Search NASA⌕ Search

SEARCH · Search NASA

Results for “human performance modeling”

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 73 records · Page 4
All sourcesNASA NTRSDOE OSTI
DOE OSTI2024

Human-Machine Shared Control for Path Following Considering Driver Fatigue Characteristics

Fatigue driving has been regarded as one of the most important factors that cause traffic accidents. This paper proposes a robust human-machine shared control strategy to improve the vehicle performance for different driver fatigue states. Firstly, the time-varying driver steering model is proposed to address the model mismatch caused by fatigue driving. And the driver fatigue evaluation system is established based on facial features to quantify driver fatigue levels. Based on the quantified fatigue levels, a novel strategy for allocating authorities of the driver and controller is developed for building the driver-vehicle interaction system. Then, to weaken the influence of parameter perturbations caused by the time-varying driver states, we design a fatigue-based shared controller through state feedback. The actuator saturation and system constraints are considered in the controller design through the robust set-invariance property to improve vehicle safety and driving comfort. The driver-in-the-loop platform is conducted to validate the effectiveness of the proposed shared steering controller. In conclusion, the experimental results show that the proposed strategy can adaptively optimize the human-machine authorities according to fatigue states and comprehensively improve vehicle performance.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2021

METHODOLOGY AND APPLICATION OF PHYSICAL SECURITY EFFECTIVENESS BASED ON DYNAMIC FORCE-ON-FORCE MODELING

This paper describes ongoing work within the Light Water Reactor Sustainability (LWRS) Program at Idaho National Laboratory (INL) to optimize security and cost of nuclear power plants (NPPs). It reviews the conservatisms in conventional physical security posture and regulations. It introduces the dynamic risk assessment tool developed at INL, Event Modeling Risk Assessment using Linked Diagrams (EMRALD). The dynamic assessment methodology leverages EMRALD to process results of force-on-force (FOF) simulations and crediting safety mitigation actions from probabilistic risk assessment (PRA) models as well as diverse and flexible coping strategies (FLEX) mitigation strategies. Timing information from these simulations are compared against the available time to perform mitigations obtained from Reactor Excursion and Leak Analysis Program (RELAP5) simulations. To illustrate the methodology, a station blackout (SBO) attack scenario was modeled in commercially available FOF simulation tools. The simulation results provide valuable insights into possible attack outcomes and as the probabilistic risk of a core damage event given these outcomes. Safety mitigation procedures were modeled in EMRALD, and were dependent on the attack outcomes by considering human operator uncertainties. RELAP5 simulations incorporating human and hardware uncertainties were performed to estimate the distribution of time-to-core damage. The results demonstrate that, even in the extreme case of a successful adversarial attack, plant mitigation strategies provide significantly high-likelihood of preventing radiological release. The proposed modeling and simulation framework of integrating FLEX equipment with FOF models enables the NPPs to credit FLEX portable equipment in the plant security posture, resulting in an efficient and optimized physical security.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗
DOE OSTI2024

Exploring the role of judgement and shared situation awareness when working with AI recommender systems

Abstract AI-advised Decision Making is a form of human-autonomy teaming in which an AI recommender system suggests a solution to a human operator, who is responsible for the final decision. This work seeks to examine the importance of judgement and shared situation awareness between humans and automated agents when interacting together in the form of a recommender systems. We propose manipulating both human judgement and shared situation awareness by providing the human decision maker with relevant information that the automated agent (AI), in the form of a recommender system, uses to generate possible courses of action. This paper presents the results of a two-phase between-subjects study in which participants and a recommender system jointly make a high-stakes decision. We varied the amount of relevant information the participant had, the assessment technique of the proposed solution, and the reliability of the recommender system. Findings indicate that this technique of supporting the human’s judgement and establishing a shared situation awareness is effective in (1) boosting the human decision maker’s situation awareness and task performance, (2) calibrating their trust in AI teammates, and (3) reducing overreliance on an AI partner. Additionally, participants were able to pinpoint the limitations and boundaries of the AI partner’s capabilities. They were able to discern situations where the AI’s recommendations could be trusted versus instances when they should not rely on the AI’s advice. This work proposes and validates a way to provide model-agnostic transparency into recommender systems that can support the human decision maker and lead to improved team performance.

Srivastava, Divya↗
DOE OSTI2020

Predicting Volume of Distribution in Humans: Performance of In Silico Methods for a Large Set of Structurally Diverse Clinical Compounds

Volume of distribution at steady state (V D,ss ) is one of the key pharmacokinetic parameters estimated during the drug discovery process. Despite considerable efforts to predict V D,ss , accuracy and choice of prediction methods remain a challenge, with evaluations constrained to a small set (<150) of compounds. To address these issues, a series of in silico methods for predicting human V D,ss directly from structure were evaluated using a large set of clinical compounds. Machine learning (ML) models were built to predict V D,ss directly and to predict input parameters required for mechanistic and empirical V D,ss predictions. In addition, log D, fraction unbound in plasma (fup), and blood-to-plasma partition ratio (BPR) were measured on 254 compounds to estimate the impact of measured data on predictive performance of mechanistic models. Furthermore, the impact of novel methodologies such as measuring partition (Kp) in adipocytes and myocytes (n = 189) on V D,ss predictions was also investigated. In predicting V D,ss directly from chemical structures, both mechanistic and empirical scaling using a combination of predicted rat and dog V D,ss demonstrated comparable performance (62%–71% within 3-fold). The direct ML model outperformed other in silico methods (75% within 3-fold, r 2 = 0.5, AAFE = 2.2) when built from a larger data set. Scaling to human from predicted V D,ss of either rat or dog yielded poor results (<47% within 3-fold). Measured fup and BPR improved performance of mechanistic V D,ss predictions significantly (81% within 3-fold, r 2 = 0.6, AAFE = 2.0). Adipocyte intracellular Kp showed good correlation to the V D,ss but was limited in estimating the compounds with low V D,ss .

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Trace Element Sampling and Partitioning Modeling to Estimate Wastewater Composition and Treatment Performance at Coal Generators

The coal combusted at coal-fired power plants (CFPPs) contains several trace elements that impact human heath if released to the environment. Effective combinations of air pollution control devices (APCDs) and water pollution control devices (WPCDs) are critical to reducing these impacts. This work sampled pulverized CFPPs owned and operated by Louisville Gas & Electricity – Kentucky Utilities (LGE-KU) during baseload and cycling conditions. Next, an open-source and easy-to-implement trace element partitioning model was developed using publicly available datasets, literature studies of trace element partitioning, and sampling data to estimate trace element partitioning in APCDs to the gas, liquid, and solid phases exiting CFPPs. These estimates of the liquid phase trace element concentration in FGD wastewater were then used to estimate trace element behavior in WPCDs and evaluate treated wastewater effluent concentrations for compliance with the Effluent Limitation Guidelines (ELG) for the Steam Electric Generating Sector. Finally, cost estimates were developed for established and emerging wastewater treatment trains to identify the most cost-effective approaches for ELG compliance.

01 COAL, LIGNITE, AND PEAT↗
DOE OSTI2023

Discovery of multi-functional polyimides through high-throughput screening using explainable machine learning

Polyimides have been widely used in modern industries because of their excellent mechanical and thermal properties, e.g., high-temperature fuel cells, displays, and aerospace composites. However, it usually takes decades of experimental efforts to develop a successful product. Aiming to expedite the discovery of high-performance polyimides, we utilize computational methods of machine learning (ML) and molecular dynamics (MD) simulations. Our study provides compelling evidence for the effectiveness of a data-driven approach in discovering novel polyimides. We first build a comprehensive library of more than 8 million hypothetical polyimides based on the polycondensation of existing dianhydride and diamine/diisocyanate molecules. Then we establish multiple ML models for the thermal and mechanical properties of polyimides based on their experimentally reported values, including glass transition temperature, Young’s modulus, and tensile yield strength. The obtained ML models demonstrate excellent predictive performance in identifying the key chemical substructures influencing the thermal and mechanical properties of polyimides. The use of explainable machine learning describes the effect of chemical substructures on individual properties, from which human experts can understand the cause of the ML model decision. Applying the well-trained ML models, we obtain property predictions of the 8 million hypothetical polyimides. Then, we screen the whole hypothetical dataset and identify three (3) best-performing novel polyimides that have better-combined properties than existing ones through Pareto frontier analysis. For an easy query of the discovered high-performing polyimides, we also create an online platform https://polyimide-explorer.herokuapp.com/ that embeds the developed ML model with interactive visualization. Furthermore, we validate the ML predictions through all-atom MD simulations and examine their synthesizability. The MD simulations are in good agreement with the ML predictions and the three novel polyimides are predicted to be easy to synthesize via Schuffenhauer’s synthetic accessibility score. Following the proposed ML guidance, we successfully synthesized a novel polyimide and the experimentally obtained high glass transition/thermal decomposition temperature demonstrated its excellent thermal stability. Here our study demonstrates an efficient way to expedite the discovery of novel polymers using ML prediction and MD validation. The high-throughput screening of a large computational dataset can serve as a general approach for new material discovery in other polymeric material exploration problems, such as organic photovoltaics, polymer membranes, and dielectrics.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗
DOE OSTI2020

Clostridioides difficile Toxin A Remodels Membranes and Mediates DNA Entry Into Cells to Activate Toll-Like Receptor 9 Signaling

Background & Aims: Clostridioides difficile toxin A (TcdA) activates the innate immune response. TcdA co-purifies with DNA. Toll-like receptor 9 (TLR9) recognizes bacterial DNA to initiate inflammation. We investigated whether DNA bound to TcdA activates an inflammatory response in murine models of C difficile infection via activation of TLR9. Methods: We performed studies with human colonocytes and monocytes and macrophages from wild-type and TLR9 knockout mice incubated with TcdA or its antagonist (ODN TTAGGG) or transduced with vectors encoding TLR9 or small-interfering RNAs. Cytokine production was measured with enzyme-linked immunosorbent assay. We studied a transduction domain of TcdA (TcdA 57-80 ), which was predicted by machine learning to have cell-penetrating activity and confirmed by synchrotron small-angle X-ray scattering. Intestines of CD1 mice, C57BL6J mice, and mice that express a form of TLR9 that is not activated by CpG DNA were injected with TcdA, TLR9 antagonist, or both. Enterotoxicity was estimated based on loop weight to length ratios. A TLR9 antagonist was tested in mice infected with C difficile. We incubated human colon explants with an antagonist of TLR9 and measured TcdA-induced production of cytokines. Results: The TcdA 57-80 protein transduction domain had membrane remodeling activity that allowed TcdA to enter endosomes. TcdA-bound DNA entered human colonocytes. TLR9 was required for production of cytokines by cultured cells and in human colon explants incubated with TcdA. TLR9 was required in TcdA-induced mice intestinal secretions and in the survival of mice infected by C difficile. Even in a protease-rich environment, in which only fragments of TcdA exist, the TcdA 57-80 domain organized DNA into a geometrically ordered structure that activated TLR9. Conclusions: TcdA from C difficile can bind and organize bacterial DNA to activate TLR9. TcdA and TcdA fragments remodel membranes, which allows them to access endosomes and present bacterial DNA to and activate TLR9. Rather than inactivating the ability of DNA to bind TLR9, TcdA appears to chaperone and organize DNA into an inflammatory, spatially periodic structure.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Bayesian chain graph models to characterize microbe-environment dynamics

Microbiome data require statistical models that can simultaneously decode microbes' reaction to the environment and interactions among microbes. While a multiresponse linear regression model seems like a straight-forward solution, we argue that treating it as a graphical model is problematic given that the regression coefficient matrix does not encode the conditional dependence structure between response and predictor nodes. This observation is especially important in biological settings when we have prior knowledge on the edges from specific experimental interventions that can only be properly encoded under a conditional dependence model. Here, we propose a chain graph model with two sets of nodes (predictors and responses) whose solution yields a graph with edges that indeed represent conditional dependence, thus agreeing with the experimenter's intuition on the average behavior of nodes under treatment. The solution to our model is sparse via the Bayesian linear regression (LASSO). In addition, we propose an adaptive extension so that different shrinkages can be applied to different edges to incorporate edge-specific prior knowledge. Our model is computationally inexpensive through an efficient Gibbs sampling algorithm and can account for binary, counting, and compositional responses via an appropriate hierarchical structure. We test the performance of our model in a variety of simulated datasets, thereby showing superior performance to state-of-the-art approaches. We further apply our model to human gut and soil microbial compositional datasets, and we highlight that CG-LASSO can estimate biologically meaningful network structures in the data.

compositional data↗
DOE OSTI2023

A dynamic parameterization of sulfuric acid–dimethylamine nucleation and its application in three-dimensional modeling

Sulfuric acid (SA) is a governing gaseous precursor for atmospheric new particle formation (NPF), a major source of global ultrafine particles, in environments studied around the world. In polluted urban atmospheres with high condensation sinks (CSs), the formation of stable SA–amine clusters, such as SA–dimethylamine (DMA) clusters, usually initializes intense NPF events. Coagulation scavenging and cluster evaporation are dominant sink processes of SA–amine clusters in urban atmospheres, yet these loss processes are not quantitatively included in the present parameterizations of SA–amine nucleation. We herein report a parameterization of SA–DMA nucleation, based on cluster dynamic simulations and quantum chemistry calculations, with certain simplifications to greatly reduce the computational costs. Compared with previous SA–DMA nucleation parameterizations, this new parameterization was able to reproduce the dependences of particle formation rates on temperature and CSs. We then incorporated it in a three-dimensional (3-D) chemical transport model to simulate the evolution of the particle number size distributions. Simulation results showed good consistency with the observations in the occurrence of NPF events and particle number size distributions in wintertime Beijing and represented a significant improvement compared to that using a parameterization without coagulation scavenging. Quantitative analysis shows that SA–DMA nucleation contributes significantly to nucleation rates and aerosol population during the 3-D simulations in Beijing (>99 % and >60 %, respectively). These results broaden the understanding of NPF in urban atmospheres and stress the necessity of including the effects of coagulation scavenging and cluster stability in simulating SA–DMA nucleation in 3-D simulations. Representing these processes is thus likely to improve model performance in particle source apportionment and quantification of aerosol effects on air quality, human health, and climate.

Li, Yuyang↗
DOE OSTI2023

Parallel evaluation of alternative skin barrier models and excised human skin for dermal absorption studies in vitro

Skin permeation is a primary consideration in the safety assessment of cosmetic ingredients, topical drugs, and human users handling veterinary medicinal products. While excised human skin (EHS) remains the ‘gold standard’ for in vitro permeation testing (IVPT) studies, unreliable supply and high cost motivate the search for alternative skin barrier models. In this study, a standardized dermal absorption testing protocol was developed to evaluate the suitability of alternative skin barrier models to predict skin absorption in humans. Under this protocol, side-by-side assessments of a commercially available reconstructed human epidermis (RhE) model (EpiDerm-200-X, MatTek), a synthetic barrier membrane (Strat-M, Sigma-Aldrich), and EHS were performed. The skin barrier models were mounted on Franz diffusion cells and the permeation of caffeine, salicylic acid, and testosterone was quantified. Transepidermal water loss (TEWL) and histology of the biological models were also compared. EpiDerm-200-X exhibited native human epidermis-like morphology, including a characteristic stratum corneum, but had an elevated TEWL as compared to EHS. The mean 6 h cumulative permeation of a finite dose (6 nmol/cm2) of caffeine and testosterone was highest in EpiDerm-200-X, followed by EHS and Strat-M. Salicylic acid permeated most in EHS, followed by EpiDerm-200-X and Strat-M. Altogether, evaluating novel alternative skin barrier models in the manner outlined herein has the potential to reduce the time from basic science discovery to regulatory impact.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

PHASE: Personalized Head-based Automatic Simulation for Electromagnetic properties in 7T MRI

Accurate and individualized human head models are becoming increasingly important for electromagnetic (EM) simulations. These simulations depend on precise anatomical representations to realistically model electric and magnetic field distributions, particularly when evaluating Specific Absorption Rate (SAR) within safety guidelines. State of the art simulations use the Virtual Population due to limited public resources and the impracticality of manually annotating patient data at scale. Here, this paper introduces Personalized Head-based Automatic Simulation for EM properties (PHASE), an automated open-source toolbox that generates high-resolution, patient-specific head models for EM simulations using paired T1-weighted (T1w) magnetic resonance imaging (MRI) and computed tomography (CT) scans with 14 tissue labels. To evaluate the performance of PHASE models, we conduct semi-automated segmentation and EM simulations on 15 real human patients, serving as the gold standard reference. The PHASE model achieved comparable global SAR and localized SAR averaged over 10 grams of tissue (SAR-10g), demonstrating its potential as a promising tool for generating large-scale human model datasets in the future. The code and models of PHASE toolbox have been made publicly available: https://github.com/hrlblab/PHASE.

Deep learning↗
DOE OSTI2023

Utility of Human In Vitro Data in Risk Assessments of Influenza A Virus Using the Ferret Model

Both in vitro and in vivo models are employed for assessing the pandemic potential of novel and emerging influenza A viruses in laboratory settings, but systematic examinations of how well viral titer measurements obtained in vitro align with results from in vivo experimentation are not frequently performed. We show that certain viral titer measurements following infection of a human bronchial epithelial cell line are positively correlated with viral titers in specimens collected from virus-inoculated ferrets and employ mathematical modeling to identify commonalities between viral infection progression between both models.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2024

Large language model evaluation for high–performance computing software development

We apply AI-assisted large language model (LLM) capabilities of GPT-3 targeting high-performance computing (HPC) kernels for (i) code generation, and (ii) auto-parallelization of serial code in C ++, Fortran, Python and Julia. Our scope includes the following fundamental numerical kernels: AXPY, GEMV, GEMM, SpMV, Jacobi Stencil, and CG, and language/programming models: (1) C++ (e.g., OpenMP [including offload], OpenACC, Kokkos, SyCL, CUDA, and HIP), (2) Fortran (e.g., OpenMP [including offload] and OpenACC), (3) Python (e.g., numpy, Numba, cuPy, and pyCUDA), and (4) Julia (e.g., Threads, CUDA.jl, AMDGPU.jl, and KernelAbstractions.jl). Kernel implementations are generated using GitHub Copilot capabilities powered by the GPT-based OpenAI Codex available in Visual Studio Code given simple + + prompt variants. To quantify and compare the generated results, we propose a proficiency metric around the initial 10 suggestions given for each prompt. For auto-parallelization, we use ChatGPT interactively giving simple prompts as in a dialogue with another human including simple “prompt engineering” follow ups. Results suggest that correct outputs for C++ correlate with the adoption and maturity of programming models. For example, OpenMP and CUDA score really high, whereas HIP is still lacking. We found that prompts from either a targeted language such as Fortran or the more general-purpose Python can benefit from adding language keywords, while Julia prompts perform acceptably well for its Threads and CUDA.jl programming models. Finally, we expect to provide an initial quantifiable point of reference for code generation in each programming model using a state-of-the-art LLM. Overall, understanding the convergence of LLMs, AI, and HPC is crucial due to its rapidly evolving nature and how it is redefining human-computer interactions.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Automated classification of big X-ray diffraction data using deep learning models

Abstract In current in situ X-ray diffraction (XRD) techniques, data generation surpasses human analytical capabilities, potentially leading to the loss of insights. Automated techniques require human intervention, and lack the performance and adaptability required for material exploration. Given the critical need for high-throughput automated XRD pattern analysis, we present a generalized deep learning model to classify a diverse set of materials’ crystal systems and space groups. In our approach, we generate training data with a holistic representation of patterns that emerge from varying experimental conditions and crystal properties. We also employ an expedited learning technique to refine our model’s expertise to experimental conditions. In addition, we optimize model architecture to elicit classification based on Bragg’s Law and use evaluation data to interpret our model’s decision-making. We evaluate our models using experimental data, materials unseen in training, and altered cubic crystals, where we observe state-of-the-art performance and even greater advances in space group classification.

Chemistry↗
DOE OSTI2021

Attend and Decode: 4D fMRI Task State Decoding Using Attention Models

Source code for Brain Attend and Decode paper. Functional magnetic resonance imaging (fMRI) is a neuroimaging modality that captures the blood oxygen level in a subject's brain while the subject either rests or performs a variety of functional tasks under different conditions. Given fMRI data, the problem of inferring the task, known as task state decoding, is challenging due to the high dimensionality (hundreds of million sampling points per datum) and complex spatio-temporal blood flow patterns inherent in the data. In this work, we propose to tackle the fMRI task state decoding problem by casting it as a 4D spatiotemporal classification problem. We present a novel architecture called Brain Attend and Decode (BAnD), that uses residual convolutional neural networks for spatial feature extraction and self-attention mechanisms for temporal modeling. We achieve significant performance gain compared to previous works on a 7-task benchmark from the large-scale Human Connectome Project-Young Adult (HCP-YA) dataset. We also investigate the transferability of BAnD's extracted features on unseen HCP tasks, either by freezing the spatial feature extraction layers and retraining the temporal model, or finetuning the entire model. The pre-trained features from BAnD are useful on similar tasks while finetuning them yields competitive results on unseen tasks/conditions.

Ng, BrendaM.↗
DOE OSTI2022

Transcriptional pathways linked to fetal and maternal hepatic dysfunction caused by gestational exposure to perfluorooctanoic acid (PFOA) or hexafluoropropylene oxide-dimer acid (HFPO-DA or GenX) in CD-1 mice

Per- and polyfluoroalkyl substances (PFAS) comprise a diverse class of chemicals used in industrial processes, consumer products, and fire-fighting foams which have become environmental pollutants of concern due to their persistence, ubiquity, and associations with adverse human health outcomes, including in pregnant persons and their offspring. Multiple PFAS are associated with adverse liver outcomes in adult humans and toxicological models, but effects on the developing liver are not fully described. Here we performed transcriptomic analyses in the mouse to investigate the molecular mechanisms of hepatic toxicity in the dam and its fetus after exposure to two different PFAS, perfluorooctanoic acid (PFOA) and its replacement, hexafluoropropylene oxide-dimer acid (HFPO-DA, known as GenX). Pregnant CD-1 mice were exposed via oral gavage from embryonic day (E) 1.5-17.5 to PFOA (0, 1, or 5 mg/kg-d) or GenX (0, 2, or 10 mg/kg-d). Maternal and fetal liver RNA was isolated (N = 5 per dose/group) and the transcriptome analyzed by Affymetrix Array. Differentially expressed genes (DEG) and differentially enriched pathways (DEP) were obtained. DEG patterns were similar in maternal liver for 5 mg/kg PFOA, 2 mg/kg GenX, and 10 mg/kg GenX (R2: 0.46-0.66). DEG patterns were similar across all 4 dose groups in fetal liver (R2: 0.59-0.81). There were more DEGs in fetal liver compared to maternal liver at the low doses for both PFOA (fetal = 69, maternal = 8) and GenX (fetal = 154, maternal = 93). Upregulated DEPs identified across all groups included Fatty Acid Metabolism, Peroxisome, Oxidative Phosphorylation, Adipogenesis, and Bile Acid Metabolism. Transcriptome-phenotype correlation analyses demonstrated > 1000 maternal liver DEGs were significantly correlated with maternal relative liver weight (R 2 >0.92). These findings show shared biological pathways of liver toxicity for PFOA and GenX in maternal and fetal livers in CD-1 mice. The limited overlap in specific DEGs between the dam and fetus suggests the developing liver responds differently than the adult liver to these chemical stressors. This work helps define mechanisms of hepatic toxicity of two structurally unique PFAS and may help predict latent consequences of developmental exposure.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2022

Comparative Assessment of Severe Acute Respiratory Syndrome Coronavirus 2 Variants in the Ferret Model

The continued spread of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) in humans necessitates evaluation of variants for enhanced virulence and transmission. We used the ferret model to perform a comparative analysis of four SARS-CoV-2 strains, including an early pandemic isolate from the United States (WA1), and representatives of the Alpha, Beta, and Delta lineages. While Beta virus was not capable of pronounced replication in ferrets, WA1, Alpha, and Delta viruses productively replicated in the ferret upper respiratory tract, despite causing only mild disease with no overt histopathological changes. Strain-specific transmissibility was observed; WA1 and Delta viruses transmitted in a direct contact setting, whereas Delta virus was also capable of limited airborne transmission. Viral RNA was shed in exhaled air particles from all inoculated animals but was highest for Delta virus. Prior infection with SARS-CoV-2 offered varied protection against reinfection with either homologous or heterologous variants. Notable genomic variants in the spike protein were most frequently detected following WA1 and Delta virus infection.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2024

SCITUNE: Aligning Large Language Models with Human-Curated Scientific Multimodal Instructions

Instruction finetuning is a popular paradigm to align large language models (LLM) with human intent. Despite its popularity, this idea is less explored in improving the LLMs to align existing foundation models with scientific disciplines, concepts and goals. In this work, we present SciTune as a tuning framework to improve the ability of LLMs to follow scientific multimodal instructions. To test our methodology, we use a human-generated scientific instruction tuning dataset and train a large multimodal model LLaMA-SciTune that connects a vision encoder and LLM for science-focused visual and language understanding. LLaMA-SciTune significantly outperforms the state-of-the-art models in the generated figure types and captions in multiple scientific multimodal benchmarks. In comparison to the models that are fine-tuned with machine generated data only, LLaMA-SciTune surpasses human performance on average and in many sub-categories on the ScienceQA benchmark.

• Artificial intelligence (AI) / machine learning ↗