Search NASA⌕ Search

SEARCH · Search NASA

Results for “text classification”

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 19 records
All sourcesNASA NTRSDOE OSTI
DOE OSTI2025

Deformable phrase level attention: A flexible approach for improving AI based medical coding

Objective: Improving the AI-driven automated medical encoding of clinical text plays a vital role in gathering information on the occurrence of diseases to improve population-level health. This work presents a novel attention mechanism designed to enhance text classification models and ensure appropriate classification of medical concepts in unstructured electronic health records. Materials and Methods: We developed a deformable, phrase-level attention mechanism to identify important lexical word-level and contextual phrase-level information from clinical text documents. We evaluated conventional and transformer-based deep learning models that we extended with our attention mechanism on the extraction of critical cancer information (e.g., site, subsite, laterality, histology, behavior) from 629,908 electronic pathology reports and on the automated medical encoding of 52,722 hospital discharge summaries. Results: Transformer-based models with the deformable, phrase-level attention mechanism achieved the best performance on the extraction of critical cancer information from pathology reports. Conventional- and transformer-based models show similar or better performance than their baseline counterparts on the automated medical encoding of clinical documents. Discussion: The addition of phrase-level information allowed models extended with our proposed method to outperform standard word-level attention. Our method showed favorable properties for the real-world application in terms of model robustness and phenotyping. These results indicate that our method is promising for automated data harmonization for common data models. Conclusion: This work proposes a novel deformable, phrase-level attention mechanism that enhances text classification models in the extraction of medical concepts from clinical text documents. We demonstrate strong performances on two clinical text datasets and showcase real-world deployability of our method.

Automated medical encoding↗
DOE OSTI2026

ZENN: A thermodynamics-inspired computational framework for heterogeneous data–driven modeling

Traditional entropy-based methods—such as cross-entropy loss in classification problems—have long been essential tools for representing the information uncertainty and physical disorder in data and for developing artificial intelligence algorithms. However, the rapid growth of data across various domains has introduced new challenges, particularly the integration of heterogeneous datasets with intrinsic disparities. To address this, we introduce a zentropy-enhanced neural network (ZENN), extending zentropy theory into the data science domain via intrinsic entropy, enabling more effective learning from heterogeneous data sources. ZENN simultaneously learns both energy and intrinsic entropy components, capturing the underlying structure of multisource data. To support this, we redesign the neural network architecture to better reflect the intrinsic properties and variability inherent in diverse datasets. We demonstrate the effectiveness of ZENN on classification tasks and energy landscape reconstructions, showing its superior generalization capabilities and robustness-particularly in predicting high-order derivatives. In image and text classification tasks, ZENN demonstrates superior generalization by introducing a learnable temperature variable that models latent multisource heterogeneity, allowing it to surpass state-of-the-art models on CIFAR-10/100, BBC News, and AG News. As a practical application in materials science, we employ ZENN to reconstruct the Helmholtz energy landscape of Fe3Pt using data generated from density functional theory and capture key material behaviors, including negative thermal expansion and the critical point in the temperature–pressure space. Overall, this work presents a zentropy-grounded framework for data-driven machine learning, positioning ZENN as a versatile and robust approach for scientific problems involving complex, heterogeneous datasets.

36 MATERIALS SCIENCE↗
DOE OSTI2026

Hybrid Quantum–Classical Graph Transformers for Efficient Sentiment Analysis

Quantum Machine Learning (QML) offers a promising paradigm that leverages quantum computing principles to develop efficient and expressive models for learning from complex and structured data. Recent advances in natural language processing (NLP) and artificial intelligence (AI) have demonstrated capabilities in understanding, generating, and reasoning over linguistic and multimodal information. In this work, we present the Quantum Graph Transformer (QGT), a hybrid quantum–classical architecture that extends graph transformer capabilities through quantum self-attention. The QGT models variable-length sentences as token graphs, where both the embedding encoding and the self-attention mechanisms are implemented using parameterized quantum circuits (PQCs), enabling efficient contextual learning with significantly fewer trainable parameters. We train QGT using both fully connected and 𝑘 -nearest-neighbor graph structures and evaluate it on five benchmark sentiment-classification datasets. Experimental results show that QGT consistently achieves higher or comparable accuracy to existing quantum NLP models and outperforms a Classical Graph Transformer (CGT) baseline with identical architecture, achieving 29.4 × fewer parameters while requiring 3–5 × fewer samples to reach comparable performance. These findings highlight the potential of graph-based quantum models as scalable and data-efficient architectures for natural language understanding.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗
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 OSTI2026

Cyote-attack Chain Estimator

Attack Chain Estimator (ACE) Application Overview The Attack Chain Estimator (ACE) Application is a sophisticated tool designed for the ingestion, classification, sequencing, and enrichment of cybersecurity threat reports. This application leverages advanced machine learning models and extensive historical data to provide comprehensive insights into cyber threats, specifically targeting Industrial Control Systems (ICS). Purpose The primary functions of the ACE Application include: Ingestion of Cybersecurity Threat Reporting: Capable of ingesting text-based threat reports in markdown or text file format. Supports ingestion of structured data from other sources in STIX/JSON format. Classification of Report’s Text-Based Events: Utilizes a DeBERTa classifier, specifically trained on cybersecurity data, to map the events to MITRE ATT&CK for ICS Tactics and Techniques. Classification is performed using multiple Jupyter notebooks and machine learning workflows hosted as FastAPI microservices: regex_data deberta_base_35_train_hft_classifier_mlflow.ipynb hft_regex_classifier_mlflow.ipynb param_train_hft_classifier_mlflow.ipynb regex_tactic_tech.ipynb Ordering of Tactics, Techniques, and Observable Events: Sequences the identified tactics, techniques, and events to form a coherent attack chain. Enrichment with Historical Attack Chain Details: Enhances the attack chain with details from historical attacks using a Markov model developed from CyOTE Precursor Analysis Report data. The Markov model is available as a FastAPI endpoint for seamless integration. Enrichment with Adversary Emulation Capabilities Data: Integrates adversary emulation capabilities data using MITRE Caldera for OT adversary abilities UUIDs. Export of Output Files: Provides options to export the enriched attack chain in JSON or CSV formats. Routing of Output to Other Applications: Facilitates routing of output to various platforms and applications, including: Threat Intelligence Platforms COREII Scout for Threat Intelligence Analysis COREII Modeling and Simulation for Adversary Emulation Technical Description The ACE Application is an advanced cybersecurity tool designed to provide detailed threat analysis and sequence generation. It is built on a robust architecture that integrates natural language processing, machine learning, and historical data modeling. Key Components: Data Ingestion Module: Handles the input of threat reports and data from various formats, ensuring flexibility in data sources. Classification Engine: Employs DeBERTa-based classifiers hosted as FastAPI microservices to analyze and classify threat report events in accordance with the MITRE ATT&CK framework for ICS. Sequence Generator: Orders the classified events into a logical attack chain, providing clear insight into the sequence of tactics and techniques used in the threat. Enrichment Engine: Integrates historical data and adversary emulation capabilities to enhance the attack chain with valuable context and additional details. The historical data enrichment is powered by a Markov model, which is available as a FastAPI endpoint. Export and Routing Module: Facilitates the export of the enriched attack chain in multiple formats and routes the output to designated applications for further analysis or emulation.

Paul, Tony [Idaho National Laboratory (INL), Idaho↗
DOE OSTI2025

Decoding substance use disorder severity from clinical notes using a large language model

Substance use disorder (SUD) poses a major concern due to its detrimental effects on health and society. SUD identification and treatment depend on a variety of factors such as severity, co-determinants (e.g., withdrawal symptoms), and social determinants of health. Existing diagnostic coding systems used by insurance providers, like the International Classification of Diseases (ICD-10), lack granularity for certain diagnoses, but American clinicians will add this granularity (as that found within the Diagnostic and Statistical Manual of Mental Disorders classification or DSM-5) as supplemental unstructured text in clinical notes. Traditional natural language processing (NLP) methods face limitations in accurately parsing such diverse clinical language. Large language models (LLMs) offer promise in overcoming these challenges by adapting to diverse language patterns. This study investigates the application of LLMs for extracting severity-related information for various SUD diagnoses from clinical notes. We propose a workflow employing zero-shot learning of LLMs with carefully crafted prompts and post-processing techniques. Through experimentation with Flan-T5, an open-source LLM, we demonstrate its superior recall compared to the rule-based approach. Focusing on 11 categories of SUD diagnoses, we show the effectiveness of LLMs in extracting severity information, contributing to improved risk assessment and treatment planning for SUD patients.

60 APPLIED LIFE SCIENCES↗
DOE OSTI2025

Leveraging large language models to automate the identification of healthcare access barriers for veterans

Objective: To develop and evaluate an automated system for identifying healthcare barriers focusing on transportation issues in veterans’ clinical notes using large language models (LLMs) and to assess the impact of different prompting strategies on classification performance and explanation consistency. Methods: We developed a hybrid system combining pattern matching for templated notes with LLM analysis for free-text notes. Using 2000 manually annotated clinical notes, we compared four prompting strategies (dual-role short, dual-role long, analysis-first, analysis-only) across Mistral-7B and Llama-3.1 models. We evaluated classification performance using standard metrics and assessed explanation consistency through embedding similarity analysis. Results: The analysis-first strategy achieved superior performance, with Mistral-7B reaching an F1 score of 0.914, outperforming traditional machine learning approaches (GBM: 0.786, BERT: 0.811). LLMs demonstrated higher explanation consistency within models (mean cosine similarity 0.887–0.908) compared to cross-model similarities (0.767–0.872). Pattern matching successfully handled 6.7% of templated notes deterministically. Mistral-7B showed greater internal consistency but higher abstention rates compared to Llama-3.1. Conclusion: Requiring LLMs to analyze evidence before classification improves both accuracy and explanation consistency for identifying transportation barriers in clinical notes. This approach enables automated barrier detection at scale while providing clinically relevant explanations, supporting both population-level healthcare planning and individual patient care decisions.

Healthcare access barriers↗
DOE OSTI2025

A knowledge-informed large language model framework for U.S. nuclear power plant shutdown initiating event classification for probabilistic risk assessment

Identifying and classifying shutdown initiating events (SDIEs) is critical for developing shutdown probabilistic risk assessment for nuclear power plants. Existing computational approaches cannot achieve satisfactory performance due to the challenges of unavailable large, labeled datasets, imbalanced event types, and label noise. To address these challenges, we propose a hybrid pipeline that integrates a knowledge-informed machine learning model to prescreen non-SDIEs and a large language model (LLM) to classify SDIEs into four types. In the prescreening stage, we proposed a set of 44 SDIE text patterns that consist of the most salient keywords and phrases from six SDIE types. Text vectorization based on the SDIE patterns generates feature vectors that are highly separable by using a simple binary classifier. The second stage builds Bidirectional Encoder Representations from Transformers (BERT)-based LLM, which learns generic English language representations from self-supervised pretraining on a large dataset and adapts to SDIE classification by fine-tuning it on an SDIE dataset. The proposed approaches are evaluated on a dataset with 10,928 events using precision, recall ratio, F 1 score, and average accuracy. In conclusion, the results demonstrate that the prescreening stage can exclude more than 97% non-SDIEs, and the LLM achieves an average accuracy of 95.1% for SDIE classification.

99 - GENERAL AND MISCELLANEOUS↗
DOE OSTI2025

Old Woman Creek Wetland Sediment and Electrochemical Sensor Microbial Community, 2023

We are developing a technique to monitor microbiological activities referred to as zero resistance ammetry, which entails the deployment of graphite electrodes in sediments. Measurement of current between electrodes of contrasting redox regimes and/or predominant terminal electron accepting processes can be used as an indicator of the extents of microbiological activity. We deployed an electrode array at depths of 2 mm, 4 mm, 76 mm, 78 mm, 152 mm, 154 mm, 227 mm, and 229 mm below the wetland sediment water interface in the Old Woman Creek National Estuarine Research Center, Huron, OH, USA (Lat. = 41.380833, Long. = -82.508889). A core was collected from adjacent sediment and subsamples were collected from depth intervals of 0 – 25 mm, 25 – 127 mm, 127 – 128 mm, and below 178 mm. To determine if the microbial communities attached to the electrodes were reflective of the adjacent sediment-associated microbial community, we conducted a 16S rRNA gene-based (V4 region) survey of these respective materials. This data package contains the results of these surveys, including metadata on the depths from which samples were collected (samples.csv), DNA extraction and sequencing information (OWC_DEPTH_AMPLICON_SEQUENCING_METADATA), sequence processing information (OWC_DEPTH_BIOINFORMATIC_METADATA.csv), an operational taxonomic unit (OTU) table (OWC_DEPTH_97OTUS_TABLE.csv), and nucleotide sequences of OTUs (OWC_DEPTH_97OTUS_SEQS.fasta). All files can be opened using a text-editing application. The fasta file is compatible with bioinformatics applications.

54 ENVIRONMENTAL SCIENCES↗
DOE OSTI2024

Semantic Stealth: Crafting Covert Adversarial Patches for Sentiment Classifiers Using Large Language Models

Deep learning models have been shown to be vulnerable to adversarial attacks, in which perturbations to their inputs cause the model to produce incorrect predictions. As opposed to adversarial attacks in computer vision, where small changes introduced to pixel values can drastically alter a model's output while remaining imperceptible to humans, text-based attacks are difficult to conceal due to the discrete nature of tokens. Consequently, unconstrained gradient-based attacks often produce adversarial examples that lack semantic meaning, rendering them detectable through visual inspection or perplexity filters. In contrast to methods that rely on gradient-based optimization in the embedding space, we propose an approach that leverages a Large Language Model's ability to generate grammatically correct and semantically meaningful text to craft adversarial patches that seamlessly blend in with the original input text. These patches can be used to alter the behavior of a target model, such as a text classifier. Since our approach does not rely on gradient backpropagation, it only requires access to the target model's confidence scores, making it a grey-box attack. We demonstrate the feasibility of our approach using open-source LLMs, including Intel's Neural Chat, Llama2, and Mistral-Instruct, to generate adversarial patches capable of altering the predictions of a distilBERT model fine-tuned on the IMDB reviews dataset for sentiment classification.

Roa Carvajal, Maria↗
DOE OSTI2024

Deep Generative Models in Energy System Applications: Review, Challenges, and Future Directions

In recent years, with the advent of mature machine learning products like ChatGPT, Stable Diffusion, and Sora, the world has witnessed tremendous changes driven by the rapid development of generative artificial intelligence (GAI). Beyond applications in text, speech, image, and video creation, deep generative models (DGMs) underpinning these cutting-edge technologies have also been employed by domain researchers to address scientific and engineering challenges. This paper aims to fill a gap in the research community by providing a systematic review of how DGMs have been utilized in energy system applications. After introducing four most popular DGMs, we review and categorize 196 research articles into five focus areas: data generation, forecasting, situational awareness, modeling, and optimal decision-making. Through this classification, we uncover trends in how DGMs are employed for each type of problem, highlighting GAI techniques that contribute to breakthroughs over traditional methods. We discuss limitations in existing literature, engineering challenges, and propose future directions, all tailored to the unique nature of problems in energy system engineering. Our goal is to offer insights for energy system domain researchers, providing a comprehensive view of existing studies and potential future opportunities.

32 ENERGY CONSERVATION, CONSUMPTION, AND UTILIZATI↗
DOE OSTI2024

A Decision Support System to Compile Environmental Mitigations from Hydropower Licensing Documents

The process of deciphering, extracting, and compiling information from texts dense with domain-specific terminology and technical jargon is a challenging endeavor. It demands considerable expertise and deep knowledge in the respective field, resulting in a labor-intensive process when executed by humans. Furthermore, the task of identifying multiple class labels in extensive texts presents a challenge due to intra- and inter-reader variability, making the process time-consuming and costly.We’re introducing a user-friendly graphical interface, fortified with a BERT model-powered decision support system. This advanced system aims to augment efficiency, curtail data collection time, and sustain high precision in data acquisition. It is instrumental in deciphering and synthesizing intricate texts teeming with a spectrum of expressions, even within similar mitigation categories. Such tasks traditionally demand substantial human effort and specialized knowledge in the domain.Our system is specifically engineered for the task of extracting environmental mitigation information to promote sustainable hydropower development from licenses issued by the Federal Energy Regulatory Commission (FERC). These license documents are comprehensive, each containing over 15,000 words and requiring the identification of 135 different class labels. We anticipate that our system will boost reading speed, improve the consistency of classification outputs among readers, and contribute to the development of a robust scientific database of environmental mitigations associated with the 2,000+ non-federal hydropower facilities licensed by FERC in the United States.

Yoon, Hong-Jun [ORNL] (ORCID:0000000254505878)↗
DOE OSTI2026

Soft symmetries of topological orders

(2+1)D topological orders possess emergent symmetries given by a group $\text{Aut}(\mathcal{C})$, which consists of the braided tensor autoequivalences of the modular tensor category $\mathcal{C}$ that describes the anyons. In this paper we discuss cases where $\text{Aut}(\mathcal{C})$ has elements that neither permute anyons nor are associated to any symmetry fractionalization but are still non-trivial, which we refer to as soft symmetries. We point out that one can construct topological defects corresponding to such exotic symmetry actions by decorating with a certain class of gauged SPT states that cannot be distinguished by their torus partition function. This gives a physical interpretation to work by Davydov on soft braided tensor autoequivalences. This has a number of important implications for the classification of gapped boundaries, non-invertible spontaneous symmetry breaking, and the general classification of symmetry-enriched topological phases of matter. Here, we also demonstrate analogous phenomena in higher dimensions, such as (3+1)D gauge theory with gauge group given by the quaternion group $Q_8$.

Gauge theory techniques↗
DOE OSTI2025

Demystifying the Resilience of Large Language Models: An End-to-End Perspective

Deep neural networks are known to be resilient to random bit-wise faults in their parameters. However, this resilience has primarily been established through evaluations of classification models. The extent to which this claim holds for large-language models remains underexplored. In this work, we conduct an extensive measurement study on the impact of random bitwise faults in commercial-scale language models. We perform an in-depth analysis of the resulting generation outputs. We first expose that these language models are not truly resilient to random bit-flips. While aggregate metrics such as accuracy may suggest resilience, an in-depth inspection of the generated outputs shows significant degradation in text quality. Our analysis also shows that tasks requiring more complex reasoning suffer more from performance and quality degradation. Moreover, we extend our analysis to models with augmented reasoning capabilities, such as Chain-of-Thought or Mixture of Experts architectures, and characterize their failure scenarios under random bit-flips.

Sun, Yu↗
DOE OSTI2026

Fungal mat growth and leaf colonization at the TRACE warming experiment, Mar - Aug 2024, Luquillo, Puerto Rico

This data package contains processed measurements on the growth of litter mat-forming fungi and the time to leaf colonization at the Tropical Responses to Altered Climate Experiment (TRACE). Located near the Sabana Field Research Station in Luquillo, Puerto Rico, the TRACE site is located in a mature, closed-canopy tropical rainforest within the Luquillo Experimental Forest (LEF). These data quantify fungal mat growth and the time to leaf colonization of fungi species Gymnopus johnstonii and Marasmius crinis-equi. The experiment was conducted in ambient (control) and experimentally warmed plots (4°C above ambient) during spring and summer periods to assess how litter mat-forming fungi respond to a range of environmental conditions of tropical wet forests. The data files include tables of relative fungal mat growth rates, time to leaf colonization, averages of soil temperature (°C), and number of dry days before leaf attachment. The data are stored in comma-separated values (CSV) format and viewable with any text editor, spreadsheet, or statistical software (e.g., R, Python, Excel). Associated metadata describe plot identifiers, measurement descriptions, and processing steps.

Agaric fungi↗
DOE OSTI2024

HarDWR - Harmonized Water Rights Records

A dataset within the Harmonized Database of Western U.S. Water Rights (HarDWR). For a detailed description of the database, please see the meta-record v2.0. Changelog v2.0 - Recalculated based on data sourced from WestDAAT - Changed using a Site ID column to identify unique records to using aa combination of Site ID and Allocation ID - Removed the Water Management Area (WMA) column from the harmonized records. The replacement is a separate file which stores the relationship between allocations and WMAs. This allows for allocations to contribute to water right amounts to multiple WMAs during the subsequent cumulative process. - Added a column describing a water rights legal status - Added "Unspecified" was a water source category - Added an acre-foot (AF) column - Added a column for the classification of the right's owner v1.02 - Added a .RData file to the dataset as a convenience for anyone exploring our code. This is an internal file, and the one referenced in analysis scripts as the data objects are already in R data objects. v1.01 - Updated the names of each file with an ID number less than 3 digits to include leading 0s v1.0 - Initial public release Description Here we present an updated database of Western U.S. water right records. This database provides consistent unique identifiers for each water right record, and a consistent categorization scheme that puts each water right record into one of seven broad use categories. These data were instrumental in conducting a study of the multi-sector dynamics of inter-sectoral water allocation changes though water markets (Grogan et al., *in review*). Specifically, the data were formatted for use as input to a process-based hydrologic model, Water Balance Model (WBM), with a water rights module (Grogan et al., *in review*). While this specific study motivated the development of the database presented here, water management in the U.S. West is a rich area of study (e.g., Anderson and Woosly, 2005; Tidwell, 2014; Null and Prudencio, 2016; Carney et al., 2021) so releasing this database publicly with documentation and usage notes will enable other researchers to do further work on water management in the U.S. West. We produced the water rights database presented here in four main steps: (1) data collection, (2) data quality control, (3) data harmonization, and (4) generation of cumulative water rights curves. Each of steps (1)-(3) had to be completed in order to produce (4), the final product that was used in the modeling exercise in Grogan et al. (*in review*). All data in each step is associated with a spatial unit called a Water Management Area (WMA), which is the unit of water right administration utilized by the state in which the right came from. Steps (2) and (3) required use to make assumptions and interpretation, and to remove records from the raw data collection. We describe each of these assumptions and interpretations below so that other researchers can choose to implement alternative assumptions an interpretation as fits their research aims. Motivation for Changing Data Sources The most significant change has been a switch from collecting the raw water rights directly from each state to using the water rights records presented in WestDAAT, a product of the Water Data Exchange (WaDE) Program under the Western States Water Council (WSWC). One of the main reasons for this is that each state of interest is a member of the WSWC, meaning that WaDE is partially funded by these states, as well as many universities. As WestDAAT is also a database with consistent categorization, it has allowed us to spend less time on data collection and quality control and more time on answering research questions. This has included records from water right sources we had previously not known about when creating v1.0 of this database. The only major downside to utilizing the WestDAAT records as our raw data is that further updates are tied to when WestDAAT is updated, as some states update their public water right records daily. However, as our focus is on cumulative water amounts at the regional scale, it is unlikely most records updates would have a significant effect on our results. The structure of WestDAAT led to several important changes to how HarWR is formatted. The most significant change is that WaDE has calculated a field known as `SiteUUID`, which is a unique identifier for the Point of Diversion (POD), or where the water is drawn from. This separate from `AllocationNativeID`, which is the identifier for the allocation of water, or the amount of water associated with the water right. It should be noted that it is possible for a single site to have multiple allocations associated with it and for an allocation to be able to be extracted from multiple sites. The site-allocation structure has allowed us to adapt a more consistent, and hopefully more realistic, approach in organizing the water right records than we had with HarDWR v1.0. This was incredibly helpful as the raw data from many states had multiple water uses within a single field within a single row of their raw data, and it was not always clear if the first water use was the most important, or simply first alphabetically. WestDAAT has already addressed this data quality issue. Furthermore, with v1.0, when there were multiple records with the same water right ID, we selected the largest volume or flow amount and disregarded the rest. As WestDAAT was already a common structure for disparate data formats, we were better able to identify sites with multiple allocations and, perhaps more importantly, allocations with multiple sites. This is particularly helpful when an allocation has sites which cross WMA boundaries, instead of just assigning the full water amount to a single WMA we are now able to divide the amount of water between the number of relevant WMAs. As it is now possible to identify allocations with water used in multiple WMAs, it is no longer practical to store this information within a single column. Instead the stAllocationToWMATab.csv file was created, which is an allocation by WMA matrix containing the percent Place of Use area overlap with each WMA. We then use this percentage to divide the allocation's flow amount between the given WMAs during the cumulation process to hopefully provide more realistic totals of water use in each area. However, not every state provides areas of water use, so like HarDWR v1.0, a hierarchical decision tree was used to assign each allocation to a WMA. First, if a WMA could be identified based on the allocation ID, then that WMA was used; typically, when available, this applied to the entire state and no further steps were needed. Second was the spatial analysis of Place of Use to WMAs. Third was a spatial analysis of the POD locations to WMAs, with the assumption that allocation's POD is within the WMA it should belong to; if an allocation still had multiple WMAs based on its POD locations, then the allocation's flow amount would be divided equally between all WMAs. The fourth, and final, process was to include water allocations which spatially fell outside of the state WMA boundaries. This could be due to several reasons, such as coordinate errors / imprecision in the POD location, imprecision in the WMA boundaries, or rights attached with features, such as a reservoir, which crosses state boundaries. To include these records, we decided for any POD which was within one kilometer of the state's edge would be assigned to the nearest WMA. Other Changes WestDAAT has Allowed In addition to a more nuanced and consistent method of assigning water right's data to WMAs, there are other benefits gained from using the WestDAAT dataset. Among those is a consistent categorization of a water right's legal status. In HarDWR v1.0, legal status was effectively ignored, which led to many valid concerns about the quality of the database related to the amounts of water the rights allowed to be claimed. The main issue was that rights with legal status' such as "application withdrawn", "non-active", or "cancelled" were included within HarDWR v1.0. These, and other water rights status' which were deemed to not be in use have been removed from this version of the database. Another major change has been the addition of the "unspecified water source category. This is water that can come from either surface water or groundwater, or the source of which is unknown. The addition of this source category brings the total number of categories to three. Due to reviewer feedback, we decided to add the acre-foot (AF) column so that the data may be more applicable to a wider audience. We added the ownerClassification column so that the data may be more applicable to a wider audience. File Descriptions The dataset is a series of various files organized by state sub-directories. In addition, each file begins with the state's name, in case the file is separate from its sub-directory for some reason. After the state name is the text which describes the contents of the file. Here is each file described in detail. Note that st is a placeholder for the state's name. stFullRecords_HarmonizedRights.csv: A file of the complete water records for each state. The column headers for each of this type of file are: state - The name of the state to which the allocations belong to. FIPS - The two digit numeric state ID code. siteID - The site location ID for POD locations. A site may have multiple allocations, which are the actual amount of water which can be drawn. In a simplified hypothetical, a farm stead may have an allocation for "irrigation" and an allocation for "domestic" water use, but the water is drawn from the same pumping equipment. It should be noted that many of the site ID appear to have been added by WaDE, and therefore may not be recognized by a given state's water rights database. allocationID - The allocation ID for the water right. For most states this is the water right ID, and what is recommended to use should a right be looked up on a given state's water rights database. The water amounts associated with these IDs tend to be finer scaled than those associated with siteID. It should be noted that some allocations may be extracted from multiple sites, particularly for larger Places of Use. ownerClassification - A classification of the types of owners for water rights. The most common is `Private` which incorporates a wide range of entities. Several classifications would be grouped into a government category, most of which are for the U.S. Federal Government. These allocations could be listed as "Federal", "United States of America", or as the names of any number of federal agencies. The last major grouping of entities is for "Native American"s. priorityDate - The date we use as the water right priority date for our modeling analysis. This is the legal priority date when it is available. However, for some rights, specifically from California and New Mexico, we used a pseudo priority date (e.g. well completion date or start of well drilling date) when a legal priority date was not available. The most questionable dates come from New Mexico, where the only date associated with certain water right records was the date the allocation was recorded in the database. As the allocation record creation tended to be within a few months of the filing of the application of the water right, from manually double checking the water rights, and our analysis focuses on aggregating water rights on the timescale of years, we determined it was acceptable to use such dates to include as many records as possible. primaryBeneficialUse - From the numerous state water use categories, WaDE categorized them into 21 categories WestDAAT. This column is the original WaDE category for the primary water use at the PoD site. allocationBeneficialUse - From the numerous state water use categories, WaDE categorized them into 21 categories for WestDAAT. This column is the original WaDE category

Economics↗
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 OSTI2025

Data from: 'Abiotic influences on continuous conifer forest structure across a subalpine watershed'

This package archives the core data used for analysis and inference in 'Abiotic influences on continuous conifer forest structure across a subalpine watershed' (Worsham et al., 2025). All data were collected in the East River, Washington Gulch, Slate River, and Coal Creek watersheds of Colorado. In the paper, we quantified the relative influence of climate, topographic, edaphic, and geologic factors on conifer stand structure and composition, and their functional relationships, at the watershed scale. We used waveform LiDAR data to derive spatially continuous stand structure metrics. We fused these with a species-level classification map to estimate tree species abundance. We applied generalized additive and generalized boosted models to evaluate the covariability of structural and compositional metrics with abiotic variables. The package contains the essential products required for reproducing our analysis and the tables and figures reported in the publication. The products comprise four classes: (1) geospatial data, (2) tabular data used for inferential analysis, (3) tabular data describing analytical results and performance statistics, and (4) a data user guide. (1) includes discretized waveform LiDAR data, locations and attributes of individual tree crowns, sampling locations and domain boundaries, a canopy height model, and raster files of estimated forest structural and compositional metrics at 100 m grid scale. (2) includes all response and explanatory variable values applied in inferential models. Response variables include conifer forest stand density, basal area, 95th percentile height, quadratic mean diameter, and others. Explanatory variables include climatic water deficit, actual evapotranspiration, elevation, heat load, soil available water content, and others. (3) includes results of training and testing several individual tree detection (ITD) algorithms, as well as inferential modeling results. (4) is a PDF user guide for this data package, including detailed descriptions and data dictionaries for all files. The data package root contains 17 assets: 8 compressed tape archive (.tar.gz) files, 5 comma-separated values (.csv) files, 3 Geographic Tagged Image File Format (GeoTIFF) (.tif) files, and 1 Portable Document Format (.pdf) file. The compressed .tar.gz archives contain ESRI shapefiles (.shp) .tif, compressed LASer (.laz), and .csv files. The archives must first be decompressed using the widely distributed command-line software utility TAR. All other files, including constituent files within the .tar.gz archives, can be opened in the open-source R statistical computing environment. Alternatively, .csv files may also be read in any simple text editor software or Microsoft Excel. Geospatial files including .shp and .tif files can also be opened in GIS software, such as QGIS (open-source) or ESRI ArcGIS (proprietary). The .pdf Data User Guide can be read with Adobe Acrobat Reader or other compatible readers.

2018 NEON and 2025 CHESS Campaigns↗