Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “tokenization”

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

The CMS DBS Query Language

The CMS experiment has implemented a flexible and powerful system enabling users to find data within the CMS physics data catalog. The Dataset Bookkeeping Service (DBS) comprises a database and the services used to store and access metadata relate d to CMS physics data. To this, we have added a generalized query system in addition to the existing web and programmatic interfaces to the DBS. This query system is based on a query language that hides the complexity of the underlying database structure by discovering the join conditions between database tables. This provides a way of querying the system that is simple and straightforward for CMS data managers and physicists to use without requiring knowledge of the database tables or keys. The DBS Query Language uses the ANTLR tool to build the input query parser and tokenizer, followed by a query builder that uses a graph representation of the DBS schema to construct the SQL query sent to underlying database. We will describe the design of the query sy stem, provide details of the language components and overview of how this component fits into the overall data discovery system architecture.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

Event driven readout architecture with non-priority arbitration for radiation detectors

A novel event driven readout architecture, EDWARD (Event Driven with Access and 8 Reset Decoder) architecture, for highly granular pixel detectors is presented. It incorporates, inter alia, an asynchronous arbitration tree based on Seitz' arbiters, removing the need for an imposed prioritization scheme. It also provides protection against glitches during readout. The system allows not only reading pixel activities, but also retrieving additional data, both analog and digital, from the pixels. A novel in-channel logic allows the entire readout process to be split into consecutive phases for additional flexibility. All operations are controlled by only one edge of the clock signal, seen as an acknowledge token, so there is no dead time between readouts.

47 OTHER INSTRUMENTATION↗

Masked Particle Modeling on Sets: Towards Self-Supervised High Energy Physics Foundation Models

Abstract We propose masked particle modeling (MPM) as a self-supervised method for learning generic, transferable, and reusable representations on unordered sets of inputs for use in high energy physics (HEP) scientific data. This work provides a novel scheme to perform masked modeling based pre-training to learn permutation invariant functions on sets. More generally, this work provides a step towards building large foundation models for HEP that can be generically pre-trained with self-supervised learning and later fine-tuned for a variety of down-stream tasks. In MPM, particles in a set are masked and the training objective is to recover their identity, as defined by a discretized token representation of a pre-trained vector quantized variational autoencoder. We study the efficacy of the method in samples of high energy jets at collider physics experiments, including studies on the impact of discretization, permutation invariance, and ordering. We also study the fine-tuning capability of the model, showing that it can be adapted to tasks such as supervised and weakly supervised jet classification, and that the model can transfer efficiently with small fine-tuning data sets to new classes and new data domains.

Heinrich, Lukas (ORCID:0000000240487584)↗

K-RET: knowledgeable biomedical relation extraction system

Abstract Motivation Relation extraction (RE) is a crucial process to deal with the amount of text published daily, e.g. to find missing associations in a database. RE is a text mining task for which the state-of-the-art approaches use bidirectional encoders, namely, BERT. However, state-of-the-art performance may be limited by the lack of efficient external knowledge injection approaches, with a larger impact in the biomedical area given the widespread usage and high quality of biomedical ontologies. This knowledge can propel these systems forward by aiding them in predicting more explainable biomedical associations. With this in mind, we developed K-RET, a novel, knowledgeable biomedical RE system that, for the first time, injects knowledge by handling different types of associations, multiple sources and where to apply it, and multi-token entities. Results We tested K-RET on three independent and open-access corpora (DDI, BC5CDR, and PGR) using four biomedical ontologies handling different entities. K-RET improved state-of-the-art results by 2.68% on average, with the DDI Corpus yielding the most significant boost in performance, from 79.30% to 87.19% in F-measure, representing a P-value of 2.91×10−12. Availability and implementation https://github.com/lasigeBioTM/K-RET.

Sousa, Diana F. (ORCID:0000000305979273)↗

Feasibility of Bell tests with the W state

The feasibility of Bell tests depends to a large extent on the tradeoff of difficulty between the preparation and the measurement of entangled systems. Polarization-entangled systems, though easy to measure, pose a relative challenge in their preparation. The opposite holds with entangled energy eigenstates for which the preparation is relatively straightforward. A way to perform measurements using a photodetector along the x, y, and z axes (in a Bloch sphere picture) on such Fock-state qubits shall be worked out, taking the W state as our entangled system. This will, by the same token, allow us to determine the minimum quantum efficiency required to perform a conclusive Bell test with the W energy eigenstate.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗

Exact solutions for a Maxwell-Kalb-Ramond action with dilaton: Localization of massless and massive modes in a sine-Gordon brane-world

We analytically find the exact propagation modes of the electromagnetic and the Kalb-Ramond fields together in a five-dimensional curved space-time. The existence and localization of gauge particles into our four-dimensional world (4D) is studied in detail on a brane-world scenario in which two gauge fields interact with a dilaton and a gravitational background. The coupling to the dilaton is different in each case causing the splitting between gauge spectra. The gauge-field zero-modes and an infinite tower of Kaluza-Klein massive states are analytically obtained. Relevant conditions on the dilaton coupling constant are found in order to identify with precision every finite tensor and vector eigenstate in the theory. An exact quantization condition on the whole mass spectrum, depending on the dilaton coupling constant and the bulk Planck mass, is inherited from the extra-dimension. This allows finding an exact rule to prevent tachyons in the theory and, by the same token, predicting a possible tensor zero-mode in 4D world. We also show that KK massive-modes contributions onto 4D physics are strongly suppressed.

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Semi-Supervised Information Extraction for Cancer Pathology Reports

Pathology reports are a main source of data for cancer surveillance programs. Manual coding of pathology reports is labor-intensive but necessary for obtaining labeled data to train automated information extraction systems. In this study, we investigated semi-supervised deep learning, improving the performance of a multitask information extraction system for automated annotation of pathology reports. We used a set of over 374,000 pathology reports from the Louisiana Tumor Registry and a novel convolutional attention-based auto-encoder. We performed a set of experiments comparing supervised training augmented with unlabeled data at 1%, 5%, 10%, and 50% of the original data size. We also compared the impact of extending text processing to include unlabeled tokens. We find that semi-supervised training consistently improved individual performance with increased micro-averaged F-scores between 0.012 and 0.064 and increased macro-averaged F-scores of up to 0.158. This demonstrates that semantic information learned via unsupervised learning can be used to improve supervised clinical task performance.

Qiu, John X.↗

Towards Automatically Matching Security Advisories to CPEs: String Similarity-based Vendor Matching

When a vulnerability is reported by the National Vulnerability Database (NVD), affected products are listed in the structured Common Platform Enumeration (CPE) format. Unfortunately, if the vulnerability is in a software library (e.g., Log4j), it will not include CPEs for each product containing that library. In these cases, security operators need to manually read the vendor's or third-party security advisories to see if their product is affected. However, these advisories do not report affected products in a structured format, which prevents automated processing, This paper makes the first effort towards automatically constructing structured CPEs for the vulnerable products in a non-NVD security advisory from the unstructured data in the advisory. Since this is a very challenging problem, this paper specifically focuses on the initial but key step of matching the un-structured vendor names in security advisories to the structured vendor representations in the standard CPE format. We explore the feasibility of using string similarity to solve the problem. The basic idea is to compare a vendor name from the non-NVD advisory with each vendor in the official CPE dictionary. The CPE vendor with the highest similarity score to the advisory's vendor will be considered as the match. We first conduct an experimental, comparative study of multiple mainstream string similarity metrics for this matching problem. To improve the performance, we then design a new string similarity metric that is adapted from an existing metric by weighing different tokens in the advisory's vendor name differently.

McClanahan, Kylie↗

Scalable Multi-Facility Workflows for Artificial Intelligence Applications in Climate Research

Earth observation satellites and earth system models are sources of vast, multi-modal datasets that are invaluable for advancing climate and environmental research. However, their scale and complexity pose significant challenges for processing and analysis. In this paper we discuss our experiences in developing and using a scientific research application using an automated multi-facility workflow that orchestrates data collection, preprocessing, artificial intelligence (AI) inferencing, and data movement across diverse computational resources, leveraging the Advanced Computing Ecosystem Testbed at the Oak Ridge Leadership Computing Facility (OLCF). We demonstrate that our workflow can be seamlessly integrated and orchestrated across research facilities managed by different federal agencies, thus allowing users to extract new scientific insights from climate datasets. The experimental results indicate that the multi-facility workflow significantly reduces processing time, enhances scalability, and maintains high efficiency across varying workloads. Notably, our workflow processes 12,000 high-resolution satellite images in just 44 seconds using 80 workers distributed across 10 nodes on the OLCF systems. Such high throughput is essential for dynamic tokenization and sharding of petascale satellite data for distributed AI model training and inferencing at scale across thousands of GPUs.

Kurihana, Takuya [ORNL] (ORCID:0000000156698565)↗

Evaluation and Demonstration of Blockchain Applicability Framework

Blockchain technology has been gaining great interest from a variety of industry sectors, including financial, food processing, and power and energy markets. Realizing the strength of blockchain technology beyond the successful application in the cryptocurrency arena, researchers have been evaluating and using blockchain for applications such as supply chain management, transactive industry (both financial and energy), system integrity, device cybersecurity, identity management, and much more. One of the unique elements of the blockchain technology that made it such a captivating technology to researchers is its plethora of features. Some of the features include smart contracts, cryptocurrency and tokenizing, immutable distributed ledger, cryptographic hashing, and digital signature. In addition, there are multiple types of blockchains, such as permissioned/private and permissionless/public, and various consensus models, such as proof-of-work, proof-of-authority, proof-of-burn, and proof-of-stake. Therefore, it is often non-trivial to determine if an application requires a blockchain. If so, what kind of blockchain and consensus is most appropriate? This paper discusses the blockchain applicability framework (BAF), which was specifically designed with the purpose to answer those questions. BAF is divided into five domains, 18 subdomains, and about 100 controls. It is designed to ingest detailed user requirements to perform a weighted evaluation that is built on mathematical constructs to determine the ideal combination of blockchain that is appropriate for an application. Along with the core logical formulation of BAF, this paper depicts the efficacy of BAF through two use cases

Gourisetti, Sri Nikhil G.↗

ASIC for High Rate 3D Position Sensitive Detectors

We report on the development of an application specific integrated circuit (ASIC) for 3D position sensitive detectors (3D PSD). The ASIC is designed to operate with pixelated wide bandgap sensors like Cadmium-Zinc-Telluride (CZT), Mercuric Iodide (Hgl2) and Thallium Bromide (TIBr). It measures the amplitudes and timings associated with an ionizing event on 128 anodes, the anode grid, and the cathode. Each channel provides low-noise charge amplification, high-order shaping with peaking time adjustable from 250 ns to 12 {micro}s, gain adjustable to 20 mV/fC or 120 mV/fC (for a dynamic range of 3.2 MeV and 530 keV in CZT), amplitude discrimination with 5-bit trimming, and positive and negative peak and timing detections. The readout can be full or sparse, based on a flag and single- or multi-cycle token passing. All channels, triggered channels only, or triggered with neighbors can be read out thus increasing the rate capability of the system to more than 10 kcps. The ASIC dissipates 330 mW which corresponds to about 2.5 mW per channel.

46 INSTRUMENTATION RELATED TO NUCLEAR SCIENCE AND ↗

ARENA: Asynchronous Reconfigurable Accelerator Ring to Enable Data-Centric Parallel Computing

The next generation HPC and data centers are likely to be reconfigurable and data-centric due to the trend of hardware specialization and the emergence of data-driven applications. In this work, we propose ARENA – an asynchronous reconfigurable accelerator ring architecture as a potential scenario on how the future HPC and data centers will be like. Despite using the coarse-grained reconfigurable arrays (CGRAs) as the substrate platform, our key contribution is not only the CGRA-cluster design itself, but also the ensemble of a new architecture and programming model that enables asynchronous tasking across a cluster of reconfigurable nodes, so as to bring specialized computation to the data rather than the reverse. We presume distributed data storage without asserting any prior knowledge on the data distribution. Hardware specialization occurs at runtime when a task finds the majority of data it requires are available at the present node. In other words, we dynamically generate specialized CGRA accelerators where the data reside. The asynchronous tasking for bringing computation to data is achieved by circulating the task token, which describes the dataflow graphs to be executed for a task, among the CGRA cluster connected by a fast ring network. Evaluations on a set of HPC and data-driven applications across different domains show that ARENA can provide better parallel scalability with reduced data movement (53.9 percent). Compared with contemporary compute-centric parallel models, ARENA can bring on average 4.37× speedup. The synthesized CGRAs and their task-dispatchers only occupy 2.93mm 2 chip area under 45nm process technology and can run at 800MHz with on average 759.8mW power consumption. ARENA also supports the concurrent execution of multi-applications, offering ideal architectural support for future high-performance parallel computing and data analytics systems.

97 MATHEMATICS AND COMPUTING↗

A Large Language Model for Determining Partial Tripping of Distributed Energy Resources

Knowing the status of individual distributed energy resources, i.e., being tripped or not, after a contingency can inform the development of an aggregated DER model. Here, this letter presents a large language model application to determine the partial tripping of distributed energy resources depending on the types, locations, and duration of faults in the transmission network. The large language model, or more specifically BERT-based approach can streamline the fault information into tokenized input, which not only reduces the complexity of the machine learning model but also demonstrates a robust performance with only limited data sets.

29 ENERGY PLANNING, POLICY, AND ECONOMY↗

Efficient Generalized Boundary Detection Using a Sliding Information Distance

In this work, we present a general machine learning algorithm for boundary detection within general signals based on an efficient, accurate, and robust approximation of the universal normalized information distance. Our approach uses an adaptive sliding information distance (SLID) combined with a wavelet-based approach for peak identification to locate the boundaries. Special emphasis is placed on developing an adaptive formulation of SLID to handle general signals with multiple unknown and/or drifting section lengths. Although specialized algorithms may outperform SLID when domain knowledge is available, these algorithms are limited to specific applications and do not generalize. SLID excels in these cases. We demonstrate the versatility and efficacy of SLID on a variety of signal types, including synthetically generated sequences of tokens, binary executables for reverse engineering applications, and time series of seismic events.

42 ENGINEERING↗

CAN-D: A Modular Four-Step Pipeline for Comprehensively Decoding Controller Area Network Data

Controller area networks (CANs) are a broadcast protocol for real-time communication of critical vehicle subsystems. Original equipment manufacturers of passenger vehicles hold secret their mappings of CAN data to vehicle signals, and these definitions vary according to make, model, and year. Without these mappings, the wealth of real-time vehicle information hidden in the CAN packets is uninterpretable, severely impeding vehicle-related research, including CAN cybersecurity and privacy studies, aftermarket tuning, efficiency and performance monitoring, and fault diagnosis to name a few. Guided by the four-part CAN signal definition, we present CAN-D (CAN-Decoder), a modular, four-step pipeline for identifying each signal's boundaries (start bit and length), endianness (byte ordering), signedness (bit-to-integer encoding), and by leveraging diagnostic standards, augmenting a subset of the extracted signals with meaningful, physical interpretation. En route to CAN-D, we provide a comprehensive review of the CAN signal reverse engineering research. All previous methods ignore endianness and signedness, rendering them incapable of decoding many standard CAN signal definitions. Incorporating endianness grows the search space from 128 to 4.72E21 signal tokenizations and introduces a web of changing dependencies. In response, we formulate, formally analyze, and provide an efficient solution to an optimization problem, allowing identification of the optimal set of signal boundaries and byte orderings. In addition, we provide two novel, state-of-the-art signal boundary classifiers—both of which are superior to previous approaches in precision and recall in three different test scenarios—and the first signedness classification algorithm, which exhibits a $>$ 97% F-score. Altogether, CAN-D is the only solution with the potential to extract any CAN signal that is also the state of the art. In evaluation on 10 vehicles of different makes, CAN-D's average $\ell ^1$ error is five times better (81% less) than all previous methods and exhibits lower average error, even when considering only signals that meet prior methods’ assumptions. Finally, CAN-D is implemented in lightweight hardware, allowing for an on-board diagnostic (OBD-II) plugin for real-time in-vehicle CAN decoding.

42 ENGINEERING↗

Attention to quantum complexity

The imminent era of error-corrected quantum computing demands robust methods to characterize quantum state complexity from limited, noisy measurements. We introduce the Quantum Attention Network (QuAN), a classical artificial intelligence (AI) framework leveraging attention mechanisms tailored for learning quantum complexity. Inspired by large language models, QuAN treats measurement snapshots as tokens while respecting permutation invariance. Combined with our parameter-efficient miniset self-attention block, this enables QuAN to access high-order moments of bit-string distributions and preferentially attend to less noisy snapshots. We test QuAN across three quantum simulation settings: driven hard-core Bose-Hubbard model, random quantum circuits, and toric code under coherent and incoherent noise. QuAN directly learns entanglement and state complexity growth from experimental computational basis measurements, including complexity growth in random circuits from noisy data. In regimes inaccessible to existing theory, QuAN unveils the complete phase diagram for noisy toric code data as a function of both noise types, highlighting AI’s transformative potential for assisting quantum hardware.

Kim, Hyejin [Cornell Univ., Ithaca, NY (United Sta↗

FORGE: Pre-Training Open Foundation Models for Science

Large language models (LLMs) are poised to revolutionize the way we conduct scientific research. However, both model complexity and pre-training cost are impeding effective adoption for the wider science community. Identifying suitable scientific use cases, finding the optimal balance between model and data sizes, and scaling up model training are among the most pressing issues that need to be addressed. In this study, we provide practical solutions for building and using LLM-based foundation models targeting scientific research use cases. We present an end-to-end examination of the effectiveness of LLMs in scientific research, including their scaling behavior and computational requirements on Frontier, the first Exascale supercomputer. We have also developed for release to the scientific community a suite of open foundation models called FORGE with up to 26B parameters using 257B tokens from over 200M scientific articles, with performance either on par or superior to other state-of-the-art comparable models. We have demonstrated the use and effectiveness of FORGE on scientific downstream tasks. Our research establishes best practices that can be applied across various fields to take advantage of LLMs for scientific discovery.

Yin, Junqi↗

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↗