Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “language 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 325 records · Page 18
All sourcesNASA NTRSDOE OSTI
DOE OSTI2023

Latent Dirichlet Allocation modeling of environmental microbiomes

Interactions between stressed organisms and their microbiome environments may provide new routes for understanding and controlling biological systems. However, microbiomes are a form of high-dimensional data, with thousands of taxa present in any given sample, which makes untangling the interaction between an organism and its microbial environment a challenge. Here we apply Latent Dirichlet Allocation (LDA), a technique for language modeling, which decomposes the microbial communities into a set of topics (non-mutually-exclusive sub-communities) that compactly represent the distribution of full communities. LDA provides a lens into the microbiome at broad and fine-grained taxonomic levels, which we show on two datasets. In the first dataset, from the literature, we show how LDA topics succinctly recapitulate many results from a previous study on diseased coral species. We then apply LDA to a new dataset of maize soil microbiomes under drought, and find a large number of significant associations between the microbiome topics and plant traits as well as associations between the microbiome and the experimental factors, e.g. watering level. This yields new information on the plant-microbial interactions in maize and shows that LDA technique is useful for studying the coupling between microbiomes and stressed organisms.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2023

Towards POI-based large-scale land use modeling: spatial scale, semantic granularity, and geographic context

The combination of spatial distribution, semantic characteristics, and sometimes temporal dynamics of POIs inside a geographic region can capture its unique land use characteristics. Most previous studies on POI-based land use modeling research focused on one geographic region and select one spatial scale and semantic granularity for land use characterization. There is a lack of understanding on the impact of spatial scale, semantic granularity, and geographic context on POI-based land use modeling, particularly large-scale land use modeling. In this study, we developed a scalable POI-based land use modeling framework and examined the impact of these three factors on POI-based land use characterization using data from three geographic regions. We developed a unified semantic representation framework for POI semantics that can help fuse heterogeneous POI data sources. Then, by combining POIs with a neural network language model, we developed a spatially explicit approach to learn the embedding representation of POIs and AOIs. We trained multiple supervised classifiers using AOI embeddings as input features to predict AOI land use at different semantic granularities. The classification performance of different land use classes was analyzed and compared across three geographic regions to identify the semantic representativeness of POI-based AOI embedding and the impact of geographic context.

58 GEOSCIENCES↗
DOE OSTI2023

Code and Data for "Quantifying Uncertainties in Greenhouse Gas Savings and Mitigation Costs with Cellulosic Biofuels"

Use of cellulosic biofuels from non-feedstocks are modeled using the BEPAM (Biofuel and Environmental Policy Analysis Model) model to quantifying the uncertainties about induced land use change effects, net greenhouse gas saving potential, and economic costs. The code is in GAMS, general algebraic modeling language. NOTE: Column 3 is titled "BAU" in "merged_BAU.gdx", "merged_RFS.gdx", and "merged_CEM.gdx", but contains "RFS" data in "merged_RFS.gdx" and "CEM" data in "merged_CEM.gdx".

BEPAM↗
DOE OSTI2025

AI Model Benchmarking for Nonproliferation Applications: Steel Thread Benchmarking Task Force Technical Report (Rev. 2)

Steel Thread is a NA-22 venture that seeks to build trustworthy, reliable AI models that can be used in a wide variety of nonproliferation tasks. A key aspect of building these models is developing appropriate benchmarks and evaluation methods, which will enable the venture to identify and adapt models to provide the most value in the nonproliferation domain. Benchmarks must be relevant to key tasks in this domain, such as question answering, information retrieval, document summarization and classification, consensus analysis, and image and data analysis. This report 1) provides an overview of benchmark design, evaluation, and challenges; 2) reviews a variety of open benchmarks, with a focus on language models and tasks; and 3) identifies benchmarks that are most relevant to Steel Thread. This report is intended to serve as a basis for further efforts to classify and evaluate benchmarks and their correlation with success on nonproliferation-specific tasks. The Steel Thread venture has defined benchmarks to be a particular combination of a dataset (or datasets) and a metric (or metrics) conceptualized as representing one or more specific tasks or sets of abilities for a specific modality. It is adopted by a research community as a shared framework for comparing methods.1 It includes 1) Data: Labeled (a designated subset not used for training, which could be all the data), 2) Metric: A way to quantify performance, 3) Task/Ability: What the benchmark is testing, 4) Protocol: A structured and repeatable evaluation process, 5) Baseline/Reference Model: For comparison; could be statistical, rule-based, SME-derived, or another model, and 6) Maintenance Plan: to update with new information over time; important for long-term utility. For further clarity, the definition includes what a benchmark, in this context, is not. It is not a corpus of training data, specific to a model (it is intended to apply to a range of models), a universal evaluation of performance, a guarantee that the ‘top’ model on the leaderboard will be the best fit for every specific use case, an all-encompassing proof of a model’s universal quality, nor is it a one-size-fits-all measure of success. It does not cover every real-world constraint (like operational, ethical, or cost considerations), a systems integration test, or a unit test. This definition was inspired by and resulted from discussions within the Steel Thread Benchmarking Task Force. This group was formed to define what we would mean as a benchmark within Steel Thread but persisted as the need to develop a thorough understanding of the large and expanding existing benchmarking space. This technical report is a result of the group’s divide and conquer approach to exploring this space. The release of benchmarks might not be progressing as quickly as model development, but it is moving very fast, as many benchmarks quickly become saturated, when state-of-the-art models score so close to the benchmark’s ceiling that their results are virtually indistinguishable. At that point, the test no longer differentiates between new systems, so researchers usually stop reporting scores as the benchmark no longer informs about improvements from the next generation of models. In the OpenAI announcement of GPT-5, they reported results on six flagship public benchmarks (AIME 2025, SWE-bench Verified, Aider Polyglot, MMMU, HealthBench Hard, GPQA) but the full system-card covers roughly thirty-five separate evaluations, comprising hundreds of test task items in total. There have been some efforts to summarize benchmarks in specific fields, like for text-to-image generation, but these surveys have had a narrow methodology scope. Therefore, a comprehensive survey of all benchmarks or even all benchmarks that could be relevant to Steel Thread is outside of the scope of this report. We chose some specific benchmarks to investigate in detail.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Third-Party Supplier Risk Re-Classification Using Multi-Model Semantic Voting and External Web Augmentation

Risk decisions in many third-party risk management (TPRM) workflows rely on static inherent risk questionnaires (IRQ). These static forms provide a snapshot of the vendor from the business users’ perspective, as these requests are processed without cross-referencing for evidence. Consequently, responses can be misinformed or embellished with inaccuracies, thereby masking the vendor’s true risk to the enterprise. This paper presents a multi-stage verification framework to augment IRQs with web evidence and a deterministic ensemble of large language model assessors to reclassify risk. In a case study of 100 submissions previously misclassified as low risk, the proposed framework correctly identified 76% of the cases as high risk, while the existing workflow identified none. McNemar’s continuity corrected statistics of 74 were obtained with a two sided p-value of 2.65 × 10-23, indicating a significantly more effective workflow compared to the legacy model.

99 - GENERAL AND MISCELLANEOUS↗
DOE OSTI2023

PRIMED for the Future: Purposing Raw Intake for Machine Learning-Enabled Detection (Final Report)

The COVID-19 pandemic demonstrated how a novel, elusive, and diffuse biological threat can engender uncertainty and misinformation, and it underscored the need for flexible analytical modalities agnostic to the identity of biological material. Yet even before the pandemic, recognition of the limitations of the current, list-based approach, which focuses on known pathogens and biotoxins, and of the importance of agent-agnostic biodetection was growing within the biosecurity community. In a 2018 report on “Biodefense in the Age of Synthetic Biology,” for example, the National Academy of Sciences stated that “an overreliance on the Select Agent List is a systematic weakness affecting many aspects of the United States’ current biodefense mitigation capability”. More recently, a group of biodefense researchers proposed the identification and adoption of “bioagent-agnostic signatures (BASs)” as a way of detecting and characterizing not only existing agents but also novel ones, an approach they believe will “enable a more flexible and resilient biodefense posture”. Indeed, the future of biodetection requires us to begin developing novel analytics that can identify anomalies and/or characteristics that indicate a potential threat, whether known or unknown, without looking for a specific signature that has been identified previously. To assess potential threats more rapidly, it is critical to develop agnostic artificial intelligence (AI)/machine learning (ML) systems that can be employed for real-time assessment of the nature and source of a perturbation. Such systems should be multi scale and multi-dimensional, integrating sensor data from a range of biological, chemical, and physical application spaces. Emerging deep learning (DL) models demonstrate exceptional promise for identification of discriminatory features within multi-dimensional datasets. DL models have the capacity to recognize and encode highly complex patterns in a wide range of input data modalities, including images, text, and biological/chemical/physical spectra. As such, they can execute a wide range of assessments and determinations that have traditionally required a human operator. The promise of advances in DL is apparent in the realm of human health and medicine. DL models have been validated for evaluating a variety of clinical threats to human health in a range of contexts, including infection and cancer, and they demonstrated improved performance in predicting stroke relative to human neurologists in some categories of data. Continuously evolving advances in AI/ML are expected to support more efficient evaluation of raw sequence, spectroscopy, and spectrometry data. For instance, recent advances and deployment of large language models (LLM) such as Generative Pre training Transformer (GPT) and Bidirectional Encoder Representations from Transformers (BERT) have already motivated application of these models for biological function prediction. As frameworks such as LLMs become larger and more complex in their representations, their capacity to serve as pre-trained models that can be fine-tuned for biological/biodetection purposes will similarly be amplified. While existing and emerging AI/ML have found broad applicability and use cases in the clinical sciences, development for environmental evaluation and biodetection has been limited. Functionalizing such capabilities for this purpose requires an understanding of the existing technical landscape and how the respective tools and algorithms are currently being employed. This landscape awareness then allows an assessment of the current practical capabilities of existing models and the anticipated requirements and development efforts that will be needed to adapt available algorithms for biodetection applications relevant to DHS. Leveraging expertise in biodetection, ML, and operational biodetection, the effort described in this report is comprised of a systematic landscape assessment (Subtask 2.1), comparative evaluation (Subtask 2.2), and formulation of a value proposition (Subtask 2.3) for the prospect of ML-enabled, agnostic biodetection from raw, or minimally-processed, datasets.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2025

CareWELL: Multimodal Region Representation Learning with Spatial Contexts for Urban Health

Rapid urbanization affects living environments by intensifying exposure to air pollution, heat, noise, and urban dynamics, which together contribute to uneven health outcomes across neighborhoods. For instance, cardiovascular, respiratory, and mental health conditions are each influenced by distinct exposures such as air pollution, extreme temperatures, or limited access to green space. These heterogeneous patterns require understanding the characteristics of geographic regions in order to explain why urban health risks vary across urban areas. Recent work in self-supervised region representation learning provides a promising way to model such characteristics from multimodal geospatial data. However, existing methods face two major limitations: (i) they often depend on non-public datasets, limiting reproducibility and applicability, and (ii) their generic pretraining objectives overlook health-relevant determinants, including temporal variability in environmental exposures and inequalities in social conditions. To address these gaps, we propose Context-Aware Region rEpresentation with Weather, Environment, and Location Learning (CareWELL). CareWELL leverages large language models to encode seasonal variability in weather, employs contrastive learning to align geo-coordinate and weather representations, and introduces a context-aware objective that integrates socio-demographic factors while preserving spatial correlations. We evaluate CareWELL by predicting six urban health outcomes in Manhattan, New York City, and demonstrate that CareWELL consistently outperforms state-of-the-art baselines as well as a traditional spatial computing method. These results suggest the importance of context-aware pretraining objectives for learning health-relevant region representations.

Namgung, Min [ORNL]↗
DOE OSTI2025

Large-scale deep learning for metastasis detection in pathology reports

Objectives No existing algorithm can reliably identify metastasis from pathology reports across multiple cancer types and the entire US population. In this study, we develop a deep learning model that automatically detects patients with metastatic cancer by using pathology reports from many laboratories and of multiple cancer types. Materials and Methods We use 60 471 unstructured pathology reports from 4 Surveillance, Epidemiology, and End Results (SEER) registries. The reports were coded into 1 of 3 labels: metastasis negative, metastases positive, or metastasis undetermined. We utilize a task-specific deep neural network trained from scratch and compare its performance with a widely used large language model (LLM). Results Our deep learning architecture trained on task-specific data outperforms a general-purpose LLM, with a recall of 0.894 compared to 0.824. We quantified model uncertainty and used it to defer reports for human review. We found that retaining 72.9% of reports increased recall from 0.894 to 0.969. Discussion A smaller deep learning architecture trained on task-specific data outperforms a general LLM. Equally critical to model performance is the incorporation of uncertainty quantification, achieved here through an abstention mechanism. Conclusions This study’s finding demonstrate the feasibility of developing algorithms to automatically identify metastatic cancer cases from unstructured pathology reports.

machine learning↗
DOE OSTI2025

Towards Content Authenticity: Multimodal Fake News Detection and AI-Generated Text Identification

In today’s digital world, the spread of fake news and the rise of AI-generated text have become major threats to content authenticity and public trust. This thesis addresses both challenges through two complementary research directions: detecting fake news using multimodal features, and identifying AI-generated text using semantic and structural reasoning. The first part of the work focuses on fake news detection by introducing a novel model that combines text and image features through a unique rotational attention mechanism. Unlike traditional attention methods, this approach rotates the roles of query, key, and value across modalities to capture deeper interactions. Additionally, the model incorporates external domain information by linking news posts to top-ranked websites from Google search results, which helps assess the credibility of content based on its broader web context. This results in a more reliable and accurate fake news detection system that outperforms existing state-of-the-art methods. The second part presents SGG-ATD, a new framework for detecting AI-generated text. It uses masked language modeling to measure sentence coherence, followed by constructing a graph where keywords—both original and predicted—are connected based on semantic and contextual similarity. A Graph Convolutional Network (GCN) is then used to learn structural relationships within the text for final classification. Experimental results demonstrate that SGG-ATD achieves high F1-scores and consistently outperforms strong baselines. This method contributes to robust AI text detection, supporting accountability and resilience against AI-driven misinformation.

Gupta, Nidhi↗
DOE OSTI2024

Gene-Metabolite Association Prediction with Interactive Knowledge Transfer Enhanced Graph for Metabolite Production

Identifying gene targets for enhancing metabolite production in metabolic engineering is challenging due to the vast research literature and the approximation in genome-scale metabolic model (GEM) simulations. Here, to address this, we propose the Gene-Metabolite Association Prediction task, which automates gene discovery for given metabolite-gene pairs, accompanied by a benchmark dataset of 2474 metabolites and 1947 genes for Saccharomyces cerevisiae (SC) and Issatchenkia orientalis (IO). This task is complicated by incomplete metabolic graphs and metabolic heterogeneity. We introduce an Interactive Knowledge Transfer mechanism based on Metabolism Graphs (IKT4Meta) to enhance prediction accuracy by integrating cross-metabolism knowledge. Using Pretrained Language Models (PLMs) to generate inter-graph links mitigates heterogeneity issues, while intra-graph links are propagated via these anchors. Gene-metabolite predictions are then performed on the enriched graphs integrating multiple microorganisms’ knowledge. Experiments show that IKT4Meta outperforms baselines by up to 12.3% in link prediction.

59 BASIC BIOLOGICAL SCIENCES↗
DOE OSTI2026

Neural Scaling Laws for Jet Generation

Recently observed empirical scaling laws describe the performance of foundation-type models as three independent key quantities -- dataset size, compute, and model parameters -- are modified. Extracting these scaling laws informs the training of large complex models for which the tuning of hyperparameters in traditional ways is not feasible. This work for the first time explores if scaling laws can also be observed for the task of particle jet generation -- both relevant as a pre-training objective for foundation models and as in-situ simulation by itself. We indeed replicate the key logarithmic scaling law behavior for model-size scaling. Beyond studying the next token prediction validation loss of the generative model, we also study the sliced Wasserstein distance of five physical quantities that are not immediately available to the model during training. Our study shows that this quantity is monotonically related to the next token prediction validation loss, meaning that this loss is indeed a good proxy for the physics performance. For the scaling with dataset size and compute, we observe substantially weaker scaling behavior of both the loss and the sliced Wasserstein distance. We analyze this behavior by introducing the concept of a learnable window, and argue that autoregressive next token prediction on jet constituents exhibits comparatively rapid saturation relative to language-model studies. We discuss possible origins of this behavior, including the stochastic nature of QCD radiation and differences between generative and supervised learning tasks in collider physics.

Amram, Oz [Fermilab]↗
DOE OSTI2025

Securing Federated Learning Against Active Reconstruction Attacks

Federated Learning (FL) has amassed notable attention for its ability to preserve user privacy while emphasizing the retainment of model training efficiency. Due to this potential, FL has been integrated in many domains, such as healthcare, finance, law, and industrial engineering, where data cannot be easily exchanged due to sensitive information and strict privacy laws. However, current research has indicated that FL protocols are easily compromised by active data reconstruction attacks employed by actively dishonest servers. The malicious modification of global model parameters allows an actively dishonest server to obtain a direct copy of users’ private data via gradient inversion. Here, this class of attacks is highly underexplored and continues to be a major challenge due to the intense threat model. In this paper, we propose OASIS as a scalable and modality-agnostic defense based on data augmentation that counteracts active data reconstruction attacks while preserving model performance. To generalize our defense, we uncover the intuition behind gradient inversion that enables these attacks and theoretically establish the conditions by which the defense can be considered robust regardless of attack design. From this, we formulate our defense with data augmentation that illustrates its ability to undermine the attack principle. We evaluate OASIS on five real-world datasets–two image-based (ImageNet and CIFAR100) and three text-based (Wikitext, Stack Overflow, and Shakespeare)–which span diverse uses cases such as vision tasks and language modeling. Comprehensive evaluations on these datasets exhibit the efficacy of OASIS and highlight its feasibility as a solution.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2023

Domain Aware Deep-learning Algorithms Integrated with Scientific-computing Technologies (DADAIST)

This technical report summarized the contribution of the DADAIST project funded by the Data Model Convergence Initiative via the Laboratory Directed Research and Development (LDRD) investments at Pacific Northwest National Laboratory (PNNL). Specifically, we report the development of the NeuroMANCER (Neural Modules with Adaptive Nonlinear Constraints and Efficient Regularizations), a new open-source Scientific Machine Learning library for formulating and solving parametric constrained optimization problems, physics-informed system identification, and parametric optimal control problems. NeuroMANCER is using differentiable programming to combine modern data-driven models and optimization modeling language into a coherent algorithmic and software framework. NeuroMANCER is a Pytorch-based framework and adopts much of its philosophy focused on research and development, rapid prototyping, and streamlined deployment. Strong emphasis is given to extensibility, interoperability with the PyTorch ecosystem, and quick adaptability to custom domain problems. Neuromancer repository contains a comprehensive library of differentiable modules, including custom activation functions, matrix factorizations, deep learning architectures, neural differential equations, differential equation solvers, implicit layers such as iterative solvers, high-level API for symbolic expressions, API for modeling and control of dynamical systems, and extensive set of tutorial code examples in the form of python scripts and jupyter notebooks.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

PVDeg: Enhancing Usability and AI-Driven Multi-Mechanism Degradation Modeling

PVDeg version 0.7.0, released in December 2025, introduced major enhancements to improve usability and performance. This update reorganized tutorials and tool notebooks to create a more intuitive experience, enabling users to easily follow and adapt workflows for their specific analyses. In addition to structural improvements, both the notebooks and core logic underwent significant optimization for efficiency, robustness, and style. These refinements were supported by new testing frameworks built on nbval and pytest, adherence to PEP8 standards, and extensive code refactoring, which collectively simplify onboarding for new developers. Looking ahead, version 0.8.0 will deliver advanced AI-driven capabilities. The primary focus is to further develop and automate the degradation workflow, designed to analyze PV module degradation across diverse locations and system configurations. By integrating large language models (LLMs) to scan literature and compile a comprehensive database of materials and degradation rates, this feature will enable modeling of multiple materials and mechanisms within a single, streamlined workflow. Users will be able to evaluate degradation impacts on different system architectures under varying environmental conditions, facilitating informed decisions on bill-of-materials optimization for specific deployment scenarios. These advancements position PVDeg as a powerful, user-friendly tool for accelerating PV reliability research and system design.

14 SOLAR ENERGY↗
DOE OSTI2026

Dual Context: Leveraging Structured Application Context for Code Generation and Runtime Feature Activation via Chat Interfaces

Integrating artificial intelligence (AI) capabilities into software applications typically involves two common paths. For developers, AI assists in generating and documenting source code and other related software engineering efforts. For users, AI assists them through question-and-answer exchanges via chatbots. Both approaches have their value, but neither effectively leverages the modularity of component-based architectures that modern web application frameworks offer. We implement a proof of concept within a centralized suite of applications used for the Atmospheric Radiation Measurement (ARM) Data Center Operational Tools, where we introduce a third integration path through the ARM Context Engine (ACE). ACE is a context driven system that uses structured contextual specifications to enable Large Language Models (LLMs) to render interactive and feature-rich user interface (UI) components directly within chat responses, alongside or in place of conventional text outputs. These specifications serve two important purposes across what we call code context and UI context. Code context provides AI-assisted development tools with structured application knowledge beyond raw code, including component relationships, architectural patterns and schematic information, enabling the generation of consistent, well-structured code. UI context defines the rules for enabling and rendering component features at runtime based on the user's natural language input, allowing end users to activate capabilities such as data export, filtering, and pagination within chat responses, without requiring code changes or redeployment. We demonstrate, through a comparative evaluation against general-purpose AI chatbots, that context-driven component rendering provides interactive capabilities that text-based responses cannot replicate, including deterministic component behavior, application-consistent design language, and on-demand feature activation. A development effort comparison further shows that features that traditionally require multi-step development cycles can be activated with a single naturallanguage request. In this ongoing work, we present ACE as an emerging approach to AI integration that positions modular, well-documented software architecture as the foundation for AI-ready applications. ACE treats context as a shared resource across both development and user-facing AI, bringing cohesion to conventionally disconnected efforts, bridging developer tooling and end-user capabilities within a single framework.

Tadimeti, Vijay [ORNL]↗
DOE OSTI2026

CAHS: Context-Aware Homology Search

Protein homology search is foundational to bioinformatics: it supports annotation transfer, structure/function inference, and evolutionary analysis over rapidly expanding sequence repositories (e.g., UniProtKB). Profile hidden Markov models (pHMMs), as implemented in HMMER, remain the most widely trusted approach because they provide statistically calibrated E-values; however, their gap behavior is fixed once a profile is trained, despite biological evidence that insertion/deletion tolerance varies across flexible loops and intrinsically disordered regions. We present CAHS (Context-Aware Homology Search), a lightweight query-time adapter for pHMM search that incorporates learned and biologically motivated signals without changing HMMER's downstream search pipeline or its calibrated E-value reporting. Given a query sequence, CAHS computes per-residue representations from a protein language model and a disorder predictor, maps these to profile coordinates, and modulates only match-state transition rows (gap-open and gap-extension probabilities) while preserving Plan7 constraints. We comprehensively evaluate CAHS across six structurally diverse protein families and multi-domain architectures against a 570k-sequence target corpus. CAHS expands detection capability, retrieving thousands of additional remote homologs at relaxed thresholds by maintaining alignment quality through flexible regions. For multi-domain proteins, context-aware modulation resolves 94% of fragmented alignments. Crucially, CAHS preserves hit-set invariance at stringent operating points (E<10-10), demonstrating increased statistical confidence without inflating false positives. Furthermore, sharper statistical distinction between homologs and background noise during early filter stages yields up to a 3.87× acceleration in end-to-end wall-clock time on high-performance computing clusters. Overall, CAHS illustrates a practical AI-for-science design pattern: augmenting a trusted probabilistic model with query-specific learned signals to improve interpretable, reproducible inference in data-rich biology.

Bhattaram, Swethasree [Georgia Institute of Techno↗
DOE OSTI2023

A GPT‐4 Reticular Chemist for Guiding MOF Discovery**

Abstract We present a new framework integrating the AI model GPT‐4 into the iterative process of reticular chemistry experimentation, leveraging a cooperative workflow of interaction between AI and a human researcher. This GPT‐4 Reticular Chemist is an integrated system composed of three phases. Each of these utilizes GPT‐4 in various capacities, wherein GPT‐4 provides detailed instructions for chemical experimentation and the human provides feedback on the experimental outcomes, including both success and failures, for the in‐context learning of AI in the next iteration. This iterative human‐AI interaction enabled GPT‐4 to learn from the outcomes, much like an experienced chemist, by a prompt‐learning strategy. Importantly, the system is based on natural language for both development and operation, eliminating the need for coding skills, and thus, make it accessible to all chemists. Our collaboration with GPT‐4 Reticular Chemist guided the discovery of an isoreticular series of MOFs, with each synthesis fine‐tuned through iterative feedback and expert suggestions. This workflow presents a potential for broader applications in scientific research by harnessing the capability of large language models like GPT‐4 to enhance the feasibility and efficiency of research activities.

Chemistry↗
DOE OSTI2025

Multi-agent AI collaboration for digital twin development and assessment

Developing a digital twin (DT) model involves different steps that encompass formulating requirements, model development, implementation, and assessment with respect to real applications. Human expertise is required to coordinate and implement different steps in the DT development and assessment process. However, certain parts of this process can be automated using artificial intelligence (AI) agents for efficient workflow development. In this work, we test and analyze a multiagent AI collaboration with humans in the loop to automate different elements of the DT development and assessment process. To implement the workflow for multiagent AI DT development and assessment, we use Autogen, a multiagent framework developed by Microsoft. Autogen offers a modular and flexible framework for configuring and designing task-specific multiagent workflows. In this framework, large language models (LLMs) form the core intelligence of the AI agents where the quality and performance of the automated element is governed by the inherent capabilities and knowledge base of the LLM. We use retrieval augmented generation to supplement the LLM with relevant domain-specific information for DT requirement formulation. We illustrate this multiagent workflow using a case study on a thermal energy storage system, focusing on how AI agents can collaborate with humans to expedite and optimize different elements of DT development and assessment process.

22 - GENERAL STUDIES OF NUCLEAR REACTORS↗