Search NASA⌕ Search

SEARCH · Search NASA

Results for “labeled data”

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 91 records · Page 5
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

VISTA

SAND2025-14753O VISTA Image Management is a tool that serves as a thin wrapper around S3 storage, enabling teams to create projects, upload images, and add labels to data. It uses standard open-source libraries, employs conventional authentication and authorization methods, and is expected to run behind a reverse proxy that handles authentication. Sandia National Laboratories is a multimission laboratory managed and operated by National Technology & Engineering Solutions of Sandia, LLC, a wholly owned subsidiary of Honeywell International Inc., for the U.S. Department of Energy’s National Nuclear Security Administration under contract DE-NA0003525.

Garland, Anthony [Sandia National Lab. (SNL-CA), L↗
DOE OSTI2024

Community Geothermal: Soil Conductivity, Borehole Design, Energy Models, and Load Data for a Residential System Development - Hinesburg, VT

This dataset contains materials from the Coalition for Community-Supported Affordable Geothermal Energy Systems (C2SAGES) project, which evaluated the techno-economic feasibility of a community geothermal system for a residential development in Hinesburg, VT. The dataset includes detailed soil conductivity test reports, energy models, borehole design reports, hourly energy loads for heating, cooling, and hot water, and design layouts. EnergyPlus was used to model building energy loads, and Modelica software was applied for geothermal loop sizing based on these loads and soil conductivity results. Python scripts for network design further refined the models. Key files include PDF reports on borehole design (with projections for 1-year, 15-year, and 30-year systems), soil conductivity test results, EnergyPlus modeling outputs, and 2D/3D design drawings in PDF, DWG, and DXF formats. Python notebooks for network design and OnePipe model files are also provided, with Modelica required for viewing certain files. Outputs and modeling data are in various formats including CSV, JPG, HTML, and IDF, with units and data clearly labeled to support understanding of system design and performance for the proposed geothermal solution.

15 GEOTHERMAL ENERGY↗
DOE OSTI2025

PDF Entity Annotation Tool (PEAT)

While different text mining approaches – including the use of Artificial Intelligence (AI) and other machine based methods - continue to expand at a rapid pace, the tools used by researchers to create the labeled datasets required for training, modeling, and evaluation remain rudimentary. Labeled datasets contain the target attributes the machine is going to learn; for example, training an algorithm to delineate between images of a car or truck would generally require a set of images with a quantitative description of the underlying features of each vehicle type. Development of labeled textual data that can be used to build natural language machine learning models for scientific literature is not currently integrated into existing manual workflows used by domain experts. Published literature is rich with important information, such as different types of embedded text, plots, and tables that can all be used as inputs to train ML/natural language processing (NLP) models, when extracted and prepared in machine readable formats. Currently, both normalized data extraction of use to domain experts and extraction to support development of ML/NLP models are labor intensive and cumbersome manual processes. Automatic extraction of data and information from formats such as PDFs that are optimized for layout and human readability, not machine readability. The PDF (Portable Document Format) Entity Annotation Tool (PEAT) was developed with the goal of allowing users to annotate publications within their current print format, while also allowing those annotations to be captured in a machine-readable format. One of the main issues with traditional annotation tools is that they require transforming the PDF into plain text to facilitate the annotation process. While doing so lessens the technical challenges of annotating data, the user loses all structure and provenance that was inherent in the underlying PDF. Also, textual data extraction from PDFs can be an error prone process. Challenges include identifying sequential blocks of text and a multitude of document formats (multiple columns, font encodings, etc.). As a result of these challenges, using existing tools for development of NLP/ML models directly from PDFs is difficult because the generated outputs are not interoperable. We created a system that allows annotations to be completed on the original PDF document structure, with no plain text extraction. The result is an application that allows for easier and more accurate annotations. In addition, by including a feature that grants the user the ability to easily create a schema, we have developed a system that can be used to annotate text for different domain-centric schemas of relevance to subject matter experts. Different knowledge domains require distinct schemas and annotation tags to support machine learning.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2024

Innovating the next generation of commercial smart building software

Nearly 30% of commercial building energy use is wasted due to equipment faults and HVAC controls problems. The result is increased emissions, compromised comfort and productivity, and less reliable coordination of building power needs with a clean grid. The energy impact alone represents $17 billion in potential savings. Today’s smart building software provides a robust solution to address these operational deficiencies. Energy management and information systems (EMIS) are saving up to 9% on average, with two-year paybacks. They are being incorporated into energy management processes, commissioning services, and utility programs. As effective as they are, two barriers prevent even deeper benefits; limited personnel to fix problems once they are identified, and the expense and time to manually implement changes in control systems. In partnership with the research community, the EMIS industry is developing new capabilities to overcome these barriers. Moving beyond siloed products for either fault detection and diagnostics, or optimal control, these new capabilities empower users to not only automatically identify faults, but also to push corrective action, and control improvements to their buildings. In this paper, several areas for enhancements are documented: ‘one-time’ correction of faults such as setpoints, schedules, and economizer lockouts; short-term active testing for automated proportional integral derivative (PID) loop tuning and functional testing; and continuous supervisory control for demand flexibility and year-round efficiency. Results are presented from a pair of partner implementations out of a dozen providers integrating these enhancements into their products, including field tests from across the country, and insights into operator acceptance and integration into operations and maintenance practices.

Casillas, Armando↗
DOE OSTI2024

Towards Agentic AI on Particle Accelerators

As particle accelerators grow in complexity, traditional control methods face increasing challenges in achieving optimal performance. This paper envisions a paradigm shift: a decentralized multi-agent framework for accelerator control, powered by Large Language Models (LLMs) and distributed among autonomous agents. We present a proposition of a self-improving decentralized system where intelligent agents handle high-level tasks and communication and each agent is specialized control individual accelerator components. This approach raises some questions: What are the future applications of AI in particle accelerators? How can we implement an autonomous complex system such as a particle accelerator where agents gradually improve through experience and human feedback? What are the implications of integrating a human-in-the-loop component for labeling operational data and providing expert guidance? We show two examples, where we demonstrate viability of such architecture.

43 PARTICLE ACCELERATORS↗
DOE OSTI2026

Synthetic data-driven deep learning for label-free autonomous atomic force microscopy

Atomic force microscopy (AFM) is a widely used tool for nanoscale characterization across materials science, energy research, and biology. However, its adoption in high-throughput materials discovery and statistically driven studies remains limited by a strong dependence on expert operator input and by the scarcity of annotated experimental AFM datasets needed to enable data-driven automation. Here, we introduce SimuScan, a synthetic-data–driven framework that enables reliable AFM feature identification, segmentation, and targeted imaging without requiring large manually labeled experimental datasets. SimuScan generates tunable, high-fidelity synthetic AFM images of defined morphologies while incorporating realistic experimental artifacts, including tip–sample convolution, noise, flattening distortions, and surface debris. These datasets are shown to support scalable, label-free training of modern deep learning models for AFM analysis. When integrated into data-driven AFM workflows, SimuScan-trained models can locate and analyze nanoscale structures across large datasets and guide targeted follow-up imaging. We validate this approach on nanostructured surfaces, DNA assemblies, and bacterial cells, demonstrating robust generalization across diverse sample types with minimal operator intervention. More broadly, this work establishes a general strategy for generating explicitly conditioned, task-relevant synthetic data to improve the reliability of downstream models in autonomous microscopy.

Millan-Solsona, Ruben [Oak Ridge National Laborato↗
DOE OSTI2026

Public Reference Data for Megawatt-Scale Hydrogen Electrolysis - NLR Historical Wind

The U.S. Department of Energy and the National Laboratory of the Rockies (NLR) demonstrate hydrogen electrolysis from variable sources, hydrogen compression and storage, and hydrogen fuel cell power production using megawatt-scale equipment at NLR’s Flatirons Campus as part of the Advanced Research on Integrated Energy Systems (ARIES) initiative. This dataset represents part of that effort and is intended for academic, national laboratory, industrial, and other stakeholders to plan, design, and validate models of megawatt-scale hydrogen technologies and diverse energy infrastructure nationwide. These data provide a baseline for how existing hydrogen electrolysis technologies perform when coupled with various energy technologies. Future datasets will demonstrate how existing hydrogen fuel cell technologies can provide controllable, dispatchable, and variable power output for artificial intelligence (AI) data centers and other variable loads. This dataset entry describes hydrogen production by conducting a statistical analysis of historical wind data over a five-year period (2020-2025) from a single 1.5MW turbine manufactured by General Electric (GE) located at NLR’s Flatirons Campus, to generate an experimental test profile that was deployed on a 1.25-MW proton exchange membrane type MC250 electrolyzer system manufactured by Nel Hydrogen . [1] While the electrolyzer balance-of-plant supports up to 2.5 MW of electrolysis, NLR only has a single 1.25-MW electrolysis stack. The historical wind data provided several metrics, however, the analysis particularly focused on the measured power output by the wind turbine. The power output time series of data for each day was categorized by total energy generation and standard deviation, and the day that represented the highest combination of these two metrics was chosen – December 25th, 2022. This process was then repeated for a moving four-hour window within this day to identify the most statistically variable period. Finally, this four-hour period was scaled by 65% to match the 1.25 MW electrolyzer. The electrolysis system controls hydrogen production by varying DC current applied to the stack, from a maximum of 3000 A to a minimum safe operation of 300 A, or 10%. Because the current – voltage characteristic changes as the stack ages and efficiency degrades, the actual minimum safe operating power changes over time. The historical wind profiles were translated from power (kilowatts) to current (amperes) using a curve fit with calibration data and sent to the electrolyzer power supply at 1 Hz frequency. For more details on the statistical analysis process, see the presentation labeled “ Public Reference Data for Megawatt-Scale Hydrogen Electrolysis” provided with each data entry. These datasets report relevant hydrogen balance-of-plant and system data, all captured at 1 Hz, including hydrogen mass production measured with an Emerson Coriolis flow meter. Each .zip file represents a single wind turbine electrolysis experiment and is formatted as follows: {technology}_{scaling factor}-{electrolyzer ramp rate in amperes/second} For instance, “wind-GE1.5MW_0.65-400.zip” represents the hour-long experiment using historical data from the wind-GE1.5MW turbine, scaled to 65%, with the electrolyzer power supply set to a maximum ramp rate (gain and slew) of 400 A/s. Each .zip folder contains the following files: A .csv file containing raw data An .xlsx file explaining all the fields in the raw data. A .png plot showing the time series of hydrogen production, electrolysis power consumption, and wind power input. A PDF file detailing the historical wind data statistical analysis used to generate the wind profile. An experiment labeled “characterization_200.zip” demonstrates the MC250 electrolyzer steady-state response with 30-minute load steps for a total duration of 5 hours. Finally, a .csv file is provided with all simulated wind experiments combined into one dataset labeled "combined_historical_wind_experiments.csv". NLR also built an AI/machine-learning predictive model based on these datasets. The model ingests the electrolyzer current command in amperes, as well as various pressures and temperatures across the system, and predicts hydrogen output in kilograms per hour. The complete model can be found at https://huggingface.co/NatLabRockies/ptmelt-hydrogen-electrolysis [1] nelhydrogen.com/product/mc-series-electrolyser .

08 HYDROGEN↗
DOE OSTI2025

Identifying Vehicle Signals in Continuous Seismic Data Using Unsupervised Machine-Learning Techniques

Seismic sensors deployed near roadways effectively capture ground vibrations generated by passing vehicles. Although both traditional and machine‐learning algorithms have been utilized for analyzing such signals, independent validation of detected vehicle events remains limited. We applied two unsupervised machine‐learning algorithms, uniform manifold approximation and projection for dimension reduction, and hierarchical density‐based spatial clustering of applications with noise, to continuous seismic data collected along a road on the main campus of Oak Ridge National Laboratory. The algorithms identified seven distinct cluster labels across the entire dataset. By comparing these cluster labels with precipitation records from a nearby weather station and image‐derived labels from a local camera system, we identified one cluster associated with rainfall and another with vehicle activity. Our algorithms identified a greater number of vehicle‐related labels compared to the camera‐derived labels because seismic data are unaffected by poor lighting conditions. The arrival times of the newly detected vehicle signals corresponded well with the road’s speed limit, supporting our findings. Our algorithm outperformed the short‐term average/long‐term average method and k‐means clustering. Our results suggest that seismic data, when analyzed with machine‐learning algorithms, can complement existing vehicle monitoring systems, particularly under challenging environmental conditions.

Chai, Chengping [Oak Ridge National Laboratory (OR↗
DOE OSTI2025

Videos, photos, and AI-derived grain size data associated with “High-throughput AI Video Surveys Enable Reproducible Multiscale Sediment Size Mapping, with Implications for Hydrobiogeochemical Parameterization”

NOTE: The manuscript associated with this data package is currently in review. The data may be revised based on reviewer feedback. Upon manuscript acceptance, this data package will be updated with the final dataset and additional metadata. This data package is associated with the manuscript “High-throughput AI Video Surveys Enable Reproducible Multiscale Sediment Size Mapping, with Implications for Hydrobiogeochemical Parameterization” under review. This data package includes five data types: 1) raw photos and videos from drone survey and walking smartphone surveys; 2) images derived from raw videos; 3) manual labeling of reference scales; 4) metadata for all images and photo resolution derived from artificial intelligence (AI) models or manual labels, 5) grain size data obtained from AI models for all photos, 6) metadata and grain size data after quality control, 7) summaries of sample efficiency for all data, and 8) computational fluid dynamics (CFD) data used to support hydro-biogeochemical (HBGC) parameter estimation. Such data is used to 1) demonstrate significant improvements in accuracy, efficiency, and quality control for grain size data collection with the help of AI models, 2) study the spatial heterogeneity of grain size and observation reproducibility based on tens of thousands of data points generated by the AI models, and 3) evaluate the impacts of grain size heterogeneity on key HBGC parameters across sediment-to-reach and hourly-to-yearly scales. In particular, the data package contains 116 folders and 179696 files. The files include 41 videos in .mov format, 64047 photos in .jpg format, 13541 video-derived photos in .png format, 12747 segmentation mask data in .tif format, 12747 segmentation data in .json format, 24771 .csv files that with metadata and grain size for each individual photo as well as water depth and velocity data from CFD and observation, 51791 .txt files of raw AI predicted labels, and 11 flight record data in .srt format. The summary for all metadata and grain size statistics information is included in “Scales_V3_NG.csv” and “Statistics_V3_NG.csv”. The summary for data that pass data quality control (QC) level 0-2 is included in “QCStatistics_V3_NG.csv”. The QC level 0 represents photos whose photo resolution is positive, excluding photos that miss reference scale. The QC level 1 means reference scale circularity uncertainty is less than 5% for smartphone images while representing photo resolution is larger than 0.44 mm/pixel for drone images. The QC level 2 means excluding photos whose grain number is less than 100, a minimum number of grains recommended by classic literature. The summary for each video’s name, length, frame rates, survey area, grain number, survey efficiency, etc. can be found in “QCSummary_V3_NG.csv”. The summary for site name, GPS coordinates, and number of images at each site can be found in “SitesSummary_V3_*.csv” files. Overall computational efficiency summary is reported in Table 4 of accompanying manuscript. Additionally, the nitrate concentration data used in this work was downloaded from an existing dataset published on ESS-DIVE (Boat-Dragged Sensor Hanford Reach.csv; Conner A. et al., 2020). We thank the United States Forest Service, Washington Department of Fish and Wildlife, Washington Department of Natural Resources, Cowiche Canyon Conservatory, Port of Benton, and the Confederated Tribes and Bands of the Yakama Nation for access to field locations where the data were collected. We also thank the Yakama Nation Tribal Council and Yakama Nation Fisheries for working with us to facilitate data collection and optimization of data usage according to their values and worldview.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2025

Self-Supervised T-GCN for Detection of Disturbance and Propagation in Power Grid

Urban power systems increasingly rely on dense sensing to monitor grid reliability, yet disturbance labels are scarce and events are rare. We present a self-supervised spatio-temporal method that detects, localizes, and characterizes grid frequency disturbances across urban areas using only unlabeled data. Our approach trains a tiny Temporal Graph Convolutional Network (T-GCN) to forecast per-site frequency residuals (deviation from 60 Hz). The sensor graph is constructed directly from signals using pre-event Pearson correlation with a cross-correlation lag penalty without geocoding. At inference, node-level anomalies are the model's forecast errors; region-level alarms arise from connected components of high-score nodes. We estimate disturbance propagation by computing per-node arrival times (first persistent exceedance), then fit a planar or time-of-arrival model to obtain direction, speed, and an epicenter proxy. With only three real events collected at decisecond resolution across U.S. cities, we evaluate the T-GCN and report time-to-detect, footprint size, and propagation consistency. We further show that short-window embeddings from the T-GCN's hidden states enable few-shot event-vs-background recognition via a simple prototypical classifier. Despite minimal data and no labels, our system yields fast, spatially coherent detection and interpretable propagation maps, offering a practical, lightweight pathway to city-scale grid resilience analytics.

Niu, Haoran [ORNL] (ORCID:0000000155228297)↗
DOE OSTI2025

Extracting Material Property Measurements from Scientific Literature with Limited Annotations

Extracting material property data from scientific text is pivotal for advancing data-driven research in chemistry and materials science; however, the extensive annotation effort required to produce training data for named entity recognition (NER) models for this task often makes it a barrier to extracting specialized data sets. Here, in this work, we present a comparative study of the conventional, supervised NER methodology to alternative few-shot learning architectures and large language model (LLM)-based approaches that mitigate the need to label large training data sets. We find that the best-performing LLM (GPT-4o) not only excels in directly extracting relevant material properties based on limited examples but also enhances supervised learning through data augmentation. We supplement our findings with error and data quality assessments to provide a nuanced understanding of factors that impact property measurement extraction.

36 MATERIALS SCIENCE↗
DOE OSTI2024

Personalized, disease-stage specific, rapid identification of immunosuppression in sepsis

Introduction Data overlapping of different biological conditions prevents personalized medical decision-making. For example, when the neutrophil percentages of surviving septic patients overlap with those of non-survivors, no individualized assessment is possible. To ameliorate this problem, an immunological method was explored in the context of sepsis. Methods Blood leukocyte counts and relative percentages as well as the serum concentration of several proteins were investigated with 4072 longitudinal samples collected from 331 hospitalized patients classified as septic (n=286), non-septic (n=43), or not assigned (n=2). Two methodological approaches were evaluated: (i) a reductionist alternative, which analyzed variables in isolation; and (ii) a non-reductionist version, which examined interactions among six (leukocyte-, bacterial-, temporal-, personalized-, population-, and outcome-related) dimensions. Results The reductionist approach did not distinguish outcomes: the leukocyte and serum protein data of survivors and non-survivors overlapped. In contrast, the non-reductionist alternative differentiated several data groups, of which at least one was only composed of survivors (a finding observable since hospitalization day 1). Hence, the non-reductionist approach promoted personalized medical practices: every patient classified within a subset associated with 100% survival subset was likely to survive. The non-reductionist method also revealed five inflammatory or disease-related stages (provisionally named ‘early inflammation, early immunocompetence, intermediary immuno-suppression, late immuno-suppression, or other’). Mortality data validated these labels: both ‘suppression’ subsets revealed 100% mortality, the ‘immunocompetence’ group exhibited 100% survival, while the remaining sets reported two-digit mortality percentages. While the ‘intermediary’ suppression expressed an impaired monocyte-related function, the ‘late’ suppression displayed renal-related dysfunctions, as indicated by high concentrations of urea and creatinine. Discussion The data-driven differentiation of five data groups may foster early and non-overlapping biomedical decision-making, both upon admission and throughout their hospitalization. This approach could evaluate therapies, at personalized level, earlier. To ascertain repeatability and investigate the dynamics of the ‘other’ group, additional studies are recommended.

Immunology↗
DOE OSTI2026

Unsupervised domain adaptation for radioisotope identification in gamma spectroscopy

Training machine learning models for radioisotope identification using gamma spectroscopy remains an elusive challenge for many practical applications, largely stemming from the difficulty of acquiring and labeling large, diverse experimental datasets. Simulations can mitigate this challenge, but the accuracy of models trained on simulated data can deteriorate substantially when deployed to an out-of-distribution operational environment. In this study, we demonstrate that unsupervised domain adaptation (UDA) can improve the ability of a model trained on synthetic data to generalize to a new testing domain, provided unlabeled data from the target domain are available. Conventional supervised techniques are unable to utilize this data because the absence of isotope labels precludes defining a supervised classification loss. Instead, we first pretrain a spectral classifier using labeled synthetic data and subsequently leverage unlabeled target data to align the learned feature representations between the source and target domains. We compare a range of different UDA techniques, finding that minimizing the maximum mean discrepancy (MMD) between source and target feature vectors yields the most consistent improvement to testing scores. For instance, using a custom transformer-based neural network, we achieved a testing accuracy of $0.904 \pm 0.022$ on an experimental LaBr test set after performing unsupervised feature alignment via MMD minimization, compared to $0.754 \pm 0.014$ before alignment. Overall, our results highlight the potential of using UDA to adapt a radioisotope classifier trained on synthetic data for real-world deployment.

Lalor, Peter W.↗
DOE OSTI2024

A Compound Data Poisoning Technique with Significant Adversarial Effects on Transformer-based Sentiment Classification Tasks

Transformer-based models have demonstrated much success in various natural language processing tasks. However, they are often vulnerable to adversarial attacks, such as data poisoning, which can intentionally fool the model into generating incorrect results. In this article, we present a novel, compound variant of a data poisoning attack on a transformer-based model that maximizes the poisoning effect while minimizing the scope of poisoning. Here we do so by combining the established data poisoning technique (label flipping) with a novel adversarial artifact selection and insertion technique aimed at minimizing detectability and the scope of the poisoning footprint. We find that by using a combination of these two techniques, we achieve a state-of-the-art attack success rate of approximately 90% while poisoning only 0.5% of the original training set, thus minimizing the scope and detectability of the poisoning action. These findings have the potential to advance the development of better data poisoning detection methods.

97 MATHEMATICS AND COMPUTING↗
DOE OSTI2026

Characterizing peak electricity demand for U.S. households: an assessment of end-use loads and demand factors

Understanding household peak electricity demand is critical to evaluate the technical need for electrical infrastructure upgrades. This study characterizes peak loads for existing and new equipment using metered data from a convenience sample of 11,940 U.S. dwellings from four sources, including 911 from two sources with end-use metering. After standardized data cleaning and labeling, we derived descriptive statistics for key metrics, such as maximum demand and demand factors, and developed predictive models relating 60- to 15-min demand for the National Electrical Code (NEC). Mean 15-min maximum demand was 9.7 kW (median 9.0 kW; IQR 7.0–11.5 kW, 95% CI 9.6–9.8 kW), indicating spare capacity in 98% of homes with hypothetical 100 A panels. Maximum demand increased with floor area and number of high-demand loads. Dwelling maximum demand was driven by higher-power, longer-duration heating appliances and vehicle charging, while most user-operated appliances contributed little. Demand factors are used to account for how most devices contribute less than their rated power to maximum demand. Existing load mean demand factors (28%; median 10%; IQR 0–58%; CI 28–29%) were higher than those for new loads (21%; median 7%; IQR 0–35%; CI 20–21%), because new loads changed the timing and magnitude of maximum demand. New high-demand loads had higher than average demand factors (40–60%). Whole dwelling demand factors support the NEC's 40% assumption, but they challenge its conservative 100% treatment of new HVAC. We propose a data-driven 50% demand factor for new equipment, which would align with metered data, improve affordability, and modernize electrical codes.

Appliances↗
DOE OSTI2024

Ice Phase Classification Made Easy with Score-Based Denoising

Accurate identification of ice phases is essential for understanding various physicochemical phenomena. However, such classification for structures simulated with molecular dynamics is complicated by the complex symmetries of ice polymorphs and thermal fluctuations. For this purpose, both traditional order parameters and data-driven machine learning approaches have been employed, but they often rely on expert intuition, specific geometric information, or large training data sets. In this work, we present an unsupervised phase classification framework that combines a score-based denoiser model with a subsequent model-free classification method to accurately identify ice phases. Further, the denoiser model is trained on perturbed synthetic data of ideal reference structures, eliminating the need for large data sets and labeling efforts. The classification step utilizes the smooth overlap of atomic position (SOAP) descriptors as the atomic fingerprint, ensuring Euclidean symmetries and transferability to various structural systems. Our approach achieves a remarkable 100% accuracy in distinguishing ice phases of test trajectories using only seven ideal reference structures of ice phases as model inputs. This demonstrates the generalizability of the score-based denoiser model in facilitating phase identification for complex molecular systems. The proposed classification strategy can be broadly applied to investigate structural evolution and phase identification for a wide range of materials, offering new insights into the fundamental understanding of water and other complex systems.

37 INORGANIC, ORGANIC, PHYSICAL, AND ANALYTICAL CH↗