Improved enzyme functional annotation prediction using contrastive learning with structural inference
Not Available
59 BASIC BIOLOGICAL SCIENCES↗
SEARCH · Engineering Papers
Results for “Model 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.
Graph Neural Network for Object Reconstruction in Liquid Argon Time Projection Chambers
The Exa.TrkX project presents a graph neural network (GNN) technique for low-level reconstruction of neutrino interactions in a Liquid Argon Time Projection Chamber (LArTPC). GNNs are still a relatively novel technique, and have shown great promise for similar reconstruction tasks in the Large Hadron Collider (LHC). Graphs describing particle interactions are formed by treating each detector hit as a node, with edges describing the relationships between hits. We utilise a multi-head attention message passing network which performs graph convolutions in order to label each node with a particle type.We present an updated variant of our GNN architecture, with several improvements. After testing the model on more realistic simulation with regions of unresponsive wires, the target was modified from edge classification to node classification in order to increase robustness. Removing edges as a classification target opens up a broader possibility space for edge-forming techniques; we explore the model’s performance across a variety of approaches, such as Delaunay triangulation, kNN, and radius-based methods. We also extend this model to the 3D context, sharing information between detector views. By using reconstructed 3D spacepoints to map detector hits from each wire plane, the model naively constructs 2D representations that are independent yet fully consistent.
46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗
popclass: A Python Package for Classifying Microlensing Events
popclass is a Python package that provides a flexible, probabilistic framework for classifying the lens of a gravitational microlensing event. Gravitational microlensing occurs when a massive foreground object (e.g., a star, white dwarf or black hole) passes in front of and lenses the light from a distant background source. This causes an apparent brightening, and shift in position, of the background source. In most cases, characteristics of the microlensing signal do not contain enough information to definitively identify the lens type. Different lens types lie in different but overlapping regions of the characteristics of the microlensing signal. For example, black holes tend to be more massive than stars and therefore cause microlensing signals that are longer. Current Galactic simulations enable us to predict where different lens types lie in the observational space and can therefore be used to classify events (e.g., Lam et al., 2020). popclass allows the user to match the characteristics of a microlensing signal with a simulation of the Galaxy to calculate lens type probabilities for the event (see Figure 1). Constraints on any microlensing signal properties and any Galactic model can be used. popclass comes with an interface to ArviZ (Kumar et al., 2019) and PyMultiNest (Buchner et al., 2014) for microlensing signal constraints, as well as pre-loaded Galactic models, plotting functionality, and methods to quantify the classification uncertainty. The probabilistic framework for popclass was developed in Perkins et al. (2024), used in Fardeen et al. (2024) and has been applied to classifying events in Kaczmarek et al. (2025).
97 MATHEMATICS AND COMPUTING↗
Mass dependence of overshooting beneath convective envelopes
State of the art stellar evolution codes [Paxton et al., 2010, Demarque et al., 2004, Weiss and Schlattl, 2008, Siess et al., 2013, Christensen-Dalsgaard, 2008] evoke a diffusive process to model convective overshooting. The diffusion coefficient [Freytag et al., 1996, Pratt et al., 2017] used in the model can be set to change with the classification of the convective zone as non-burning, H-burning, He-burning, or metal-burning. It can also be set to change at defined evolutionary points, such as the bottom of the asymptotic giant branch, or during the third dredge up [Herwig, 2000, Lugaro et al., 2003]. Aside from these abrupt changes, the diffusion coefficient is typically locked to a percentage of the pressure scale height measured at the convective boundary. However, there is no theoretical reason for convective overshooting, or indeed other convective properties, to change in the same way that the pressure scale height changes as a star evolves.
79 ASTRONOMY AND ASTROPHYSICS↗
Figure Descriptive Text Extraction Using Ontological Representation
Abstract Experimental research publications provide figure form resources including graphs, charts, and any type of images to effectively support and convey methods and results. To describe figures, authors add captions, which are often incomplete, and more descriptions reside in body text. This work presents a method to extract figure descriptive text from the body of scientific articles. We adopted ontological semantics to aid concept recognition of figure-related information, which generates human- and machine-readable knowledge representations from sentences. Our results show that conceptual models bring an improvement in figure descriptive sentence classification over word-based approaches.
97 MATHEMATICS AND COMPUTING↗
Machine Learning for DUNE Supernova Trigger
One of the major scientific goals of the Deep Underground Neutrino Experiment (DUNE) is to detect and measure the neutrino flux originating from galactic core-collapse supernovae. These neutrinos provide an opportunity to study the end of life evolution of massive stars, and reveal information about the structure of core-collapse that is not visible in observations of the electromagnetic spectrum. Because of the rarity of these events, it is crucial that DUNE is able to detect supernova neutrino interactions when they occur. However, this will require sifting through a large quantity of data, motivating the development of a trigger algorithm to identify significant events and discard irrelevant data. Machine learning provides a potential approach to building this trigger. This project generates ADC and ground truth images of simulated neutrino interactions in a LArTPC detector to be used for machine learning, and uses them to train a sparse Convolutional Neural Network (C NN). The performance of this model when applied to the task of pixel classification based on interaction type is examined. This project found that the sparse CNN approach has the potential to have high accuracy in pixel classification, meaning it may be highly relevant to the development of a supernova neutrino trigger for the DUNE far detector.
Damish, S.↗
How efficiently can AI recognize Wireless Devices?
This poster presents a hardware benchmarking methodology for a 3-layer CNN waveform classifier deployed using ONNX Runtime on an NVIDIA Jetson AGX Orin. The dataset consist of 9 signal types, -30 to +30 dB SNR with 5dB increments. Benchmarking on the Jetson AGX Orin gave an accuracy of 91.9% and GPU throughput of 107,120 predictions/sec (23× faster than CPU). The Jetson GPU reached approximately 27M samples/sec with stable performance but fell below the 40 MHz rate needed for real-time radio feeds. Sustained testing of 5 minutes confirmed stable performance with no memory leaks, establishing a reproducible benchmarking baseline for future edge-deployment optimization.
99 - GENERAL AND MISCELLANEOUS↗
Kinetics-Informed Neural Networks
Chemical kinetics and reaction engineering consists of the phenomenological framework for the disentanglement of reaction mechanisms, optimization of reaction performance and the rational design of chemical processes. Here, we utilize feed-forward artificial neural networks as basis functions to solve ordinary differential equations (ODEs) constrained by differential algebraic equations (DAEs) that describe microkinetic models (MKMs). We present an algebraic framework for the mathematical description and classification of reaction networks, types of elementary reaction, and chemical species. Under this framework, we demonstrate that the simultaneous training of neural nets and kinetic model parameters in a regularized multi-objective optimization setting leads to the solution of the inverse problem through the estimation of kinetic parameters from synthetic experimental data. We analyze a set of scenarios to establish the extent to which kinetic parameters can be retrieved from transient kinetic data, and assess the robustness of the methodology with respect to statistical noise. Furthermore, this approach to inverse kinetic ODEs can assist in the elucidation of reaction mechanisms based on transient data.
36 MATERIALS SCIENCE↗
First Impressions: Early-time Classification of Supernovae Using Host-galaxy Information and Shallow Learning
Substantial effort has been devoted to the characterization of transient phenomena from photometric information. Automated approaches to this problem have taken advantage of complete phase coverage of an event, limiting their use for triggering rapid follow-up of ongoing phenomena. In this work, we introduce a neural network with a single recurrent layer designed explicitly for early photometric classification of supernovae (SNe). Our algorithm leverages transfer learning to account for model misspecification, host-galaxy photometry to solve the data-scarcity problem soon after discovery, and a custom weighted loss to prioritize accurate early classification. We first train our algorithm using state-of-the-art transient and host-galaxy simulations, then adapt its weights and validate it on the spectroscopically confirmed SNe Ia, SNe II, and SNe Ib/c from the Zwicky Transient Facility Bright Transient Survey. On observed data, our method achieves an overall accuracy of 82% ± 2% within 3 days of an event’s discovery, and an accuracy of 87% ± 5% within 30 days of discovery. At both early and late phases, our method achieves comparable or superior results to the leading classification algorithms with a simpler network architecture. These results help pave the way for rapid photometric and spectroscopic follow-up of scientifically valuable transients discovered in massive synoptic surveys.
79 ASTRONOMY AND ASTROPHYSICS↗
Chaconne: A Statistical Approach to Nonlocal Compression for Supervised Learning, Semi-Supervised Learning, and Anomaly Detection
This project developed a novel statistical understanding of compression analytics (CA), which has challenged and clarified some core assumptions about CA, and enabled the development of novel techniques that address vital challenges of national security. Specifically, this project has yielded the development of novel capabilities including 1. Principled metrics for model selection in CA, 2. Techniques for deriving/applying optimal classification rules and decision theory to supervised CA, including how to properly handle class imbalance and differing costs of misclassification, 3. Two techniques for handling nonlocal information in CA, 4. A novel technique for unsupervised CA that is agnostic with regard to the underlying compression algorithm, 5. A framework for semisupervised CA when a small number of labels are known in an otherwise large unlabeled dataset. 6. The academic alliance component of this project has focused on the development of a novel exemplar-based Bayesian technique for estimating variable length Markov models (closely related to PPM [prediction by partial matching] compression techniques). We have developed examples illustrating the application of our work to text, video, genetic sequences, and unstructured cybersecurity log files.
99 GENERAL AND MISCELLANEOUS↗
Sentiment analysis of the United States public support of nuclear power on social media using large language models
This study utilized large language models (LLMs) to analyze public sentiment in the United States (US) regarding nuclear power on social media, focusing on X/Twitter, considering climate change challenges and advancements in nuclear power technology. Approximately, 1.26 million nuclear tweets from 2008–2023 were examined to fine-tune LLMs for sentiment classification. We found the crucial role of accurate data labeling for model performance, with potential implications for a 15% improvement, achieved through high-confidence labels. LLMs demonstrated better performance compared to traditional machine learning classifiers, with reduced susceptibility to overfitting and up to 96% classification accuracy. LLMs are used to segment the US public tweets into policy and energy-related categories, revealing that 68% are politically themed. Policy tweets tended to convey negative sentiment, often reflecting opposing political perspectives and focusing on nuclear deals and international relations. Energy-related tweets covered diverse topics with predominantly neutral to positive sentiment, indicating broad support for nuclear power in 48 out of 50 US states. The US public positive sentiments toward nuclear power stemmed from its high power density, reliability regardless of weather conditions, environmental benefits, application versatility, and recent innovations and advancements in both fission and fusion technologies. Negative sentiments primarily focused on waste management, high capital costs, and safety concerns. The neutral campaign highlighted global nuclear facts and advancements, with varying tones leaning towards positivity or negativity. An interesting neutral theme was the advocacy for the combined use of renewable and nuclear energy to attain net-zero goals.
Energy & Fuels↗
HydraGNN_OPF_GFM_2026 - Ensemble of predictive graph foundation models for power grid applications
This dataset supports research on graph foundation models for optimal power flow (OPF) on electric grids using HydraGNN. It contains heterogeneous graph representations of PGLib-OPF cases spanning systems from 14 to 13,659 buses, together with packed HDF5 datasets for pretraining, feasibility classification, and N-1 contingency analysis. The release includes OPF solution data, downstream fine-tuning datasets, pretrained HeteroSAGE and HeteroHEAT model checkpoints, hyperparameter-optimization summaries across multiple heterogeneous GNN architectures, and aggregated fine-tuning results for sample-efficiency studies. The dataset is designed to enable scalable training, evaluation, and transfer-learning studies for OPF surrogate modeling, including node-level AC-OPF solution prediction, graph-level prediction, feasibility classification, operating-condition generalization, and contingency-response tasks.
24 POWER TRANSMISSION AND DISTRIBUTION↗
Evaluating Deception Detection Model Robustness To Linguistic Variation
With the increasing use of automated, machine learning-driven tools and the downstream impact that algorithmic judgements can have, it is critical to develop models that are robust to evolving or manipulated inputs. Evaluating the reliability of multimodal models across linguistic variations to understand model susceptibility to intentional linguistic adversarial attacks as well as natural linguistic variations is essential in this pursuit. We present extensive analysis of model robustness and susceptibility to linguistic variations in the setting of deceptive news detection, a difficult classification task that is an increasingly important problem to solve with the impact of misinformation spread online. We evaluate the effectiveness of incorporating adversarial defense strategies and measure model susceptibility to state-of-the-art adversarial attacks using two types of linguistic attacks — character and word perturbations. We consider two multiclass prediction tasks — a 3-way classification of tweets as trustworthy, propaganda, or disinformation; and a 4-way classification as clickbait, hoax, satire, or conspiracy — and compare the performance of three embeddings that have been state-of-the-art for several NLP tasks — GloVe, ELMo, and BERT — to highlight consistent trends in susceptibility, high confidence misclassifications, and high impact failures. We find that character or mixed ensemble models are the most effective defense mechanisms and that character perturbations are a more effective attack than word perturbations for deception classification.
adversarial evaluation↗
Superconducting Radio-frequency Cavity Fault Classification Using Machine Learning at Jefferson Laboratory
We report on the development of machine learning models for classifying C100 superconducting radiofrequency (SRF) cavity faults in the Continuous Electron Beam Accelerator Facility (CEBAF) at Jefferson Lab. Of the 418 SRF cavities in CEBAF, 96 are designed with a digital low-level RF system configured such that a cavity fault triggers recordings of RF signals for each of eight cavities in the cryomodule. Subject matter experts analyze the collected time-series data and identify which of the eight cavities faulted first and classify the type of fault. This information is used to find trends and strategically deploy mitigations to problematic cryomodules. However, manually labeling the data is laborious and time-consuming. By leveraging machine learning, near real-time - rather than postmortem - identification of the offending cavity and classification of the fault type has been implemented. We discuss the performance of the machine learning models during a recent physics run. We also discuss efforts for further insights into fault types through unsupervised learning techniques and present preliminary work on cavity and fault prediction using data collected prior to a failure event.
Tennant, C. D.↗
Air classification of forest residue for tissue and ash separation efficiency
The goal of this Case Study was to evaluate the performance of air classification of logging residues toward meeting conversion CMAs for carbon and ash contents, as compared to the static status quo Base Case system in which the residues are first dried and then ground in a hammer mill with a 6 mm screen and fines < 1.18 mm are removed. Also considered were moisture and ash impacts on throughput and Overall Operating Effectiveness (OOE), as well as delivered feedstock cost and minimum fuel selling price (MFSP). Laboratory data on the impacts of fan speed and moisture content on the separation efficiency of soil ash, needles and bark from white wood were received from FCIC Subtask 5.2. Average throughput and energy consumption data were obtained from the Bioenergy Feedstock National User Facility (BFNUF) for the same air classifier. These data were utilized to develop the necessary response surface equations to perform throughput analysis using discrete event simulation. Because the Base Case status quo system utilizes drying prior to grinding, we modeled the Case Study with drying prior to air classification and subsequent grinding of the separated white wood to isolate the individual quality and cost impacts of air classification relative to the Base Case system. Key takeaways from this Case Study are that air classification improves the quality of the final material, but increases the production cost, especially when lights are disposed; this becomes a tradeoff between increased conversion yield and the extra cost. Removing material should be done as early in the process as possible. Each operation that occurs prior to removing the material increases the cost of the disposed material and leads to wasted energy expenditures. As processes are included or modified, the impact on monetary and energy cost should be included in the decision process. Identifying alternative uses and the associated value for material that is separated from the feedstock stream will have a significant impact on the delivered cost of the material. Although there was an increase in production cost by adding air classification, there was a resulting benefit to the MFSP when meeting or exceeding all of the conversion CMAs; this was an important finding that should be explored further with FCIC Subtask 8.3 and potentially FCIC Task 6.
09 BIOMASS FUELS↗
An Evaluation of Representation Learning Methods in Particle Physics Foundation Models
We present a systematic evaluation of representation learning objectives for particle physics within a unified framework. Our study employs a shared transformer-based particle-cloud encoder with standardized preprocessing, matched sampling, and a consistent evaluation protocol on a jet classification dataset. We compare contrastive (supervised and self-supervised), masked particle modeling, and generative reconstruction objectives under a common training regimen. In addition, we introduce targeted supervised architectural modifications that achieve state-of-the-art performance on benchmark evaluations. This controlled comparison isolates the contributions of the learning objective, highlights their respective strengths and limitations, and provides reproducible baselines. We position this work as a reference point for the future development of foundation models in particle physics, enabling more transparent and robust progress across the community.
Chen, Michael [Caltech]↗
Air Classification of Forest Residue for Tissue and Ash Separation Efficiency
The goal of this Case Study was to evaluate the performance of air classification of logging residues toward meeting conversion CMAs for carbon and ash contents, as compared to the static status quo Base Case system in which the residues are first dried and then ground in a hammer mill with a 6 mm screen and fines less than 1.18 mm are removed. Also considered were moisture and ash impacts on throughput and Overall Operating Effectiveness (OOE), as well as delivered feedstock cost and minimum fuel selling price (MFSP). Laboratory data on the impacts of fan speed and moisture content on the separation efficiency of soil ash, needles and bark from white wood were received from FCIC Subtask 5.2: Preprocessing, High Temperature Conversion Preprocessing (Jordan Klinger and Tiasha Bhattacharjee, INL). Average throughput and energy consumption data were obtained from the Bioenergy Feedstock National User Facility (BFNUF) (Neal Yancey, INL) for the same air classifier. These data were utilized to develop the necessary response surface equations to perform throughput analysis using discrete event simulation. Feedstock-Conversion Interface Consortium. Because the Base Case status quo system utilizes drying prior to grinding, we modeled the Case Study with drying prior to air classification and subsequent grinding of the separated white wood to isolate the individual quality and cost impacts of air classification relative to the Base Case system.
CMA↗
Key Technical Issues for Greater-Than-Class-C (GTCC) Waste Disposal - 20164
Greater-than-Class C (GTCC) waste is low-level waste (LLW) that exceeds the Class C concentrations tabulated in Title 10, Code of Federal Regulations (CFR), Part 61. Disposal of GTCC waste in the near surface (i.e. upper 30 m of the earth's surface) is prohibited in the United States. Though GTCC waste disposal is generally prohibited, the Commission can approve disposal according to 10 CFR 61.55(a)(2)(iv) on a case-by-case basis. The waste classification tables, developed in the early 1980's, were based on model projections of dose to an inadvertent intruder in agricultural, construction, and discovery exposure scenarios. The assumptions and parameters used were documented in a series of public reports. Though different disposal facility designs and site conditions were considered, the waste classification tables were based on a disposal facility design located in a specific environment. The tables were based on shallow (i.e., top few meters) trench facility designs and did not consider deeper facilities. To determine the suitability of GTCC waste disposal in the near-surface, site-specific analyses must account for differences between GTCC waste and Class A, B, and C LLW (hereafter, traditional LLW). GTCC waste can have concentrations of radionuclides that are much higher than traditional LLW. Because of these higher concentrations, processes that are typically not significant for traditional LLW may be significant with respect to disposal of GTCC waste. These processes include, but are not limited to, heat generation, criticality, and radiolysis. The form of the waste as well as the barriers to release of the waste (e.g. waste package) could be substantially different than they are for traditional LLW. These barriers need to be considered when assessing the impacts of accidents during receipt and placement and in evaluating long-term performance. Some GTCC radioactivity is either embedded in stainless steel or contained in stainless steel barriers. Stainless steel can have very low corrosion rates under a variety of environmental conditions. Finally, and possibly most importantly, GTCC waste would likely need to be disposed deeper than traditional LLW to reduce the probability of disturbance. The waste classification table values of traditional LLW are based on an inadvertent intruder excavating into the waste and bringing some of the material to the land surface. If waste is deeper than approximately 5 m, the excavation scenario becomes very unlikely. Therefore, other intruder scenarios, such as drilling exposure scenarios, need to be evaluated. This paper summarizes key technical issues for the disposal of GTCC waste. Our previous study shows that certain GTCC waste may be suitable for near surface disposal whereas others may not. This study may support ongoing technical analyses assessing potential disposal of GTCC in a near surface disposal facility. (authors)
12 MANAGEMENT OF RADIOACTIVE AND NON-RADIOACTIVE W↗