arXiv subjects

Youngkyu Kim

Publications and source records attributed to Youngkyu Kim.

8 recordsLinked to original sources

arXiv2026

An Explainable Approach to Document-level Translation Evaluation with Topic Modeling

The advent of NMT has expanded the scope of translation beyond isolated sentences, enabling context to be preserved across paragraphs and documents. However, current evaluation metrics largely remain restricted to the sentence level and typically depend on reference translations. Without references, existing metrics cannot provide a clear basis for their quality assessments. To address these limitations, we propose an evaluation framework that independently extracts and compares latent topic structures within source and translated texts. This framework utilises various topic modelling techniques, including LSA, LDA and BERTopic, to achieve this. Our methodology captures statistical frequency information and semantic context, providing a comprehensive evaluation of the entire document. It aligns key topic tokens across languages using a bilingual dictionary and quantifies thematic consistency via cosine similarity. This allows us to evaluate how faithfully the translation maintains the thematic integrity of the source text, even in the absence of reference translations. To this end, we used a large scale dataset of 9.38 million Korean to English sentence pairs from AI Hub, which includes pre evaluated BLEU scores. We also calculated CometKiwi, a state of the art, reference free metric for this dataset, in order to conduct a comparative analysis with our proposed, topic based framework. Through this analysis, we confirmed that, unlike existing metrics, our framework evaluates the differentiated attribute of document level thematic units. Furthermore, visualising the key tokens that underpin the quantitative evaluation score provides clear insight into translation quality. Consequently, this study contributes to effectively complementing the existing translation evaluation system by proposing a new metric that intuitively identifies whether the document's theme has been preserved.

cs.CE↗

arXiv2025

Defining Foundation Models for Computational Science: A Call for Clarity and Rigor

The widespread success of foundation models in natural language processing and computer vision has inspired researchers to extend the concept to scientific machine learning and computational science. However, this position paper argues that as the term "foundation model" is an evolving concept, its application in computational science is increasingly used without a universally accepted definition, potentially creating confusion and diluting its precise scientific meaning. In this paper, we address this gap by proposing a formal definition of foundation models in computational science, grounded in the core values of generality, reusability, and scalability. We articulate a set of essential and desirable characteristics that such models must exhibit, drawing parallels with traditional foundational methods, like the finite element and finite volume methods. Furthermore, we introduce the Data-Driven Finite Element Method (DD-FEM), a framework that fuses the modular structure of classical FEM with the representational power of data-driven learning. We demonstrate how DD-FEM addresses many of the key challenges in realizing foundation models for computational science, including scalability, adaptability, and physics consistency. By bridging traditional numerical methods with modern AI paradigms, this work provides a rigorous foundation for evaluating and developing novel approaches toward future foundation models in computational science.

cs.LG↗

arXiv2025

Evaluation of Thermal Control Based on Spatial Thermal Comfort with Reconstructed Environmental Data

Achieving thermal comfort while maintaining energy efficiency is a critical objective in building system control. Conventional thermal comfort models, such as the Predicted Mean Vote (PMV), rely on both environmental and personal variables. However, the use of fixed-location sensors limits the ability to capture spatial variability, which reduces the accuracy of occupant-specific comfort estimation. To address this limitation, this study proposes a new PMV estimation method that incorporates spatial environmental data reconstructed using the Gappy Proper Orthogonal Decomposition (Gappy POD) algorithm. In addition, a group PMV-based control framework is developed to account for the thermal comfort of multiple occupants. The Gappy POD method enables fast and accurate reconstruction of indoor temperature fields from sparse sensor measurements. Using these reconstructed fields and occupant location data, spatially resolved PMV values are calculated. Group-level thermal conditions are then derived through statistical aggregation methods and used to control indoor temperature in a multi-occupant living lab environment. Experimental results show that the Gappy POD algorithm achieves an average relative error below 3\% in temperature reconstruction. PMV distributions varied by up to 1.26 scale units depending on occupant location. Moreover, thermal satisfaction outcomes varied depending on the group PMV method employed. These findings underscore the importance for adaptive thermal control strategies that incorporate both spatial and individual variability, offering valuable insights for future occupant-centric building operations.

cs.CE↗

arXiv2025

Real-Time Reconstruction of Ground Motion During Small Magnitude Earthquakes: A Pilot Study

This study presents a pilot investigation into a novel method for reconstructing real-time ground motion during small magnitude earthquakes (M < 4.5), removing the need for computationally expensive source characterization and simulation processes to assess ground shaking. Small magnitude earthquakes, which occur frequently and can be modeled as point sources, provide ideal conditions for evaluating real-time reconstruction methods. Utilizing sparse observation data, the method applies the Gappy Auto-Encoder (Gappy AE) algorithm for efficient field data reconstruction. This is the first study to apply the Gappy AE algorithm to earthquake ground motion reconstruction. Numerical experiments conducted with SW4 simulations demonstrate the method's accuracy and speed across varying seismic scenarios. The reconstruction performance is further validated using real seismic data from the Berkeley area in California, USA, demonstrating the potential for practical application of real-time earthquake data reconstruction using Gappy AE. As a pilot investigation, it lays the groundwork for future applications to larger and more complex seismic events.

physics.geo-ph↗

arXiv2024

Gappy AE: A Nonlinear Approach for Gappy Data Reconstruction using Auto-Encoder

We introduce a novel data reconstruction algorithm known as Gappy auto-encoder (Gappy AE) to address the limitations associated with Gappy proper orthogonal decomposition (Gappy POD), a widely used method for data reconstruction when dealing with sparse measurements or missing data. Gappy POD has inherent constraints in accurately representing solutions characterized by slowly decaying Kolmogorov N-widths, primarily due to its reliance on linear subspaces for data prediction. In contrast, Gappy AE leverages the power of nonlinear manifold representations to address data reconstruction challenges of conventional Gappy POD. It excels at real-time state prediction in scenarios where only sparsely measured data is available, filling in the gaps effectively. This capability makes Gappy AE particularly valuable, such as for digital twin and image correction applications. To demonstrate the superior data reconstruction performance of Gappy AE with sparse measurements, we provide several numerical examples, including scenarios like 2D diffusion, 2D radial advection, and 2D wave equation problems. Additionally, we assess the impact of four distinct sampling algorithms - discrete empirical interpolation method, the S-OPT algorithm, Latin hypercube sampling, and uniformly distributed sampling - on data reconstruction accuracy. Our findings conclusively show that Gappy AE outperforms Gappy POD in data reconstruction when sparse measurements are given.

physics.comp-ph↗

arXiv2020

Efficient space-time reduced order model for linear dynamical systems in Python using less than 120 lines of code

A classical reduced order model (ROM) for dynamical problems typically involves only the spatial reduction of a given problem. Recently, a novel space-time ROM for linear dynamical problems has been developed, which further reduces the problem size by introducing a temporal reduction in addition to a spatial reduction without much loss in accuracy. The authors show an order of a thousand speed-up with a relative error of less than 0.00001 for a large-scale Boltzmann transport problem. In this work, we present for the first time the derivation of the space-time Petrov-Galerkin projection for linear dynamical systems and its corresponding block structures. Utilizing these block structures, we demonstrate the ease of construction of the space-time ROM method with two model problems: 2D diffusion and 2D convection diffusion, with and without a linear source term. For each problem, we demonstrate the entire process of generating the full order model (FOM) data, constructing the space-time ROM, and predicting the reduced-order solutions, all in less than 120 lines of Python code. We compare our Petrov-Galerkin method with the traditional Galerkin method and show that the space-time ROMs can achieve O(100) speed-ups with O(0.001) to O(0.0001) relative errors for these problems. Finally, we present an error analysis for the space-time Petrov-Galerkin projection and derive an error bound, which shows an improvement compared to traditional spatial Galerkin ROM methods.

math.NA↗

arXiv2020

Efficient nonlinear manifold reduced order model

Traditional linear subspace reduced order models (LS-ROMs) are able to accelerate physical simulations, in which the intrinsic solution space falls into a subspace with a small dimension, i.e., the solution space has a small Kolmogorov n-width. However, for physical phenomena not of this type, such as advection-dominated flow phenomena, a low-dimensional linear subspace poorly approximates the solution. To address cases such as these, we have developed an efficient nonlinear manifold ROM (NM-ROM), which can better approximate high-fidelity model solutions with a smaller latent space dimension than the LS-ROMs. Our method takes advantage of the existing numerical methods that are used to solve the corresponding full order models (FOMs). The efficiency is achieved by developing a hyper-reduction technique in the context of the NM-ROM. Numerical results show that neural networks can learn a more efficient latent space representation on advection-dominated data from 2D Burgers' equations with a high Reynolds number. A speed-up of up to 11.7 for 2D Burgers' equations is achieved with an appropriate treatment of the nonlinear terms through a hyper-reduction technique.

math.NA↗

arXiv2020

A fast and accurate physics-informed neural network reduced order model with shallow masked autoencoder

Traditional linear subspace reduced order models (LS-ROMs) are able to accelerate physical simulations, in which the intrinsic solution space falls into a subspace with a small dimension, i.e., the solution space has a small Kolmogorov n-width. However, for physical phenomena not of this type, e.g., any advection-dominated flow phenomena, such as in traffic flow, atmospheric flows, and air flow over vehicles, a low-dimensional linear subspace poorly approximates the solution. To address cases such as these, we have developed a fast and accurate physics-informed neural network ROM, namely nonlinear manifold ROM (NM-ROM), which can better approximate high-fidelity model solutions with a smaller latent space dimension than the LS-ROMs. Our method takes advantage of the existing numerical methods that are used to solve the corresponding full order models. The efficiency is achieved by developing a hyper-reduction technique in the context of the NM-ROM. Numerical results show that neural networks can learn a more efficient latent space representation on advection-dominated data from 1D and 2D Burgers' equations. A speedup of up to 2.6 for 1D Burgers' and a speedup of 11.7 for 2D Burgers' equations are achieved with an appropriate treatment of the nonlinear terms through a hyper-reduction technique. Finally, a posteriori error bounds for the NM-ROMs are derived that take account of the hyper-reduced operators.

math.NA↗