arXiv subjects

Seungmin Lee

Publications and source records attributed to Seungmin Lee.

At least 19 recordsLinked to original sources

arXiv2026

When the Knowledge Base Becomes the Gold Standard: Measuring Resource-Shared Evaluation Loops in Entity-Level Machine Translation

The Seungjeongwon Ilgi, a UNESCO Memory of the World record, is only 37.4% translated, and the most conspicuous failure mode in automatic translation is the person name -- a misread name corrupts the historical fact rather than merely the surface. Low-resource historical domains have no expert gold standard for entity translation, so practitioners substitute a knowledge base (KB) for the gold. That KB is the same resource injected into the system: scoring becomes self-referential and the metric measures instruction compliance rather than translation quality. We measure this loop. Using expert person-name annotations from the National Institute of Korean History as a gold independent of the injection pipeline, we hold the entity set fixed and vary only the provenance of the correct reading. Of 527 expert-annotated mentions, only 31.1% lie outside the injection pipeline, and the residual loop is not uniform -- in the overlapping segment the injected reading agrees with the human translation 97.8% of the time against 70.1% in the independent one, so the segment that looks healthiest is the one the loop is holding up. Across four models, a difference-in-differences analysis shows the gain from KB injection is confined to the segment whose gold shares the injected resource; in the independent segment it is at or below zero. Post-injection preservation clusters in a narrow 0.910-0.996 band even though baseline capability differs fivefold, so the reported gain is the complement of prior performance and weaker models appear to improve more dramatically. On an independent sample built by removing the construction filter, the measure replicates within model (overlapping intervals) while discriminating between models (non-overlapping intervals) -- it reflects a property of the model, not of the sample.

cs.CL↗

arXiv2026

Automated Terminal-to-Housing Assembly System for Flat Ribbon Cable Harness

This paper presents a sensor-minimal automated assembly system for bidirectional single-row flat ribbon cable harnesses (FRCHs). Unlike conventional peg-in-hole or single-terminal insertion tasks, FRCH assembly involves mechanically coupled multi-terminal insertion under flexible and dense geometric constraints. To address this problem, the proposed system performs the assembly through a purely mechanical sequence consisting of Cable Alignment, Lean & Slide, Weaving, and Clamping, without relying on active sensing or vision. Each mechanism is designed to progressively reduce correlated terminal misalignment, insertion interference, and instability before final locking. Experiments on bidirectional single-row FRCHs achieved an 83.75% end-to-end process success rate over 80 trials, with success rates of 85.0% and 82.5% in the first and second halves, respectively. The cycle time was 33 s under half-speed operation. To the best of our knowledge, this work presents the first automated prototype for FRCH terminal-to-housing assembly for multi-pin housings.

cs.RO↗

arXiv2026

Sagnac-Loop-Reflector Fabry-Perot Lattices for Modular 1D Topological Photonics

We introduce a modular silicon-photonic Fabry-Perot resonator lattice based on cascaded tunable Sagnac loop reflectors. Each SLR is controlled by a single directional-coupler cross-coupling coefficient, enabling modular control of the effective lattice hoppings. As a representative example, alternating two SLR types maps the lattice onto the Su-Schrieffer-Heeger model in the weak-coupling limit. We derive the Bloch dispersion via a transfer-matrix formulation and obtain an effective tight-binding Hamiltonian in the weak-coupling limit. S-parameter simulations of a 20-site lattice show an isolated midgap resonance with edge-localized power profiles in the topological phase, and disorder tests show robustness against symmetry-preserving hopping perturbations. Our results establish SLR-based FP lattices as a complementary platform for on-chip topological photonics.

physics.optics↗

arXiv2026

REZE: Representation Regularization for Domain-adaptive Text Embedding Pre-finetuning

Recent text embedding models are often adapted to specialized domains via contrastive pre-finetuning (PFT) on a naive collection of scattered, heterogeneous tasks. However, this approach often introduces task-induced bias alongside domain knowledge, leading to uncontrolled representation shifts that distort the pretrained embedding geometry and cause substantial performance degradation. To address this issue, we propose REZE, a representation regularization framework that explicitly controls representation shift during embedding pre-finetuning. REZE operates on the relations of anchor-positive pairs and decomposes them in an eigenspace. It then measures task-wise dispersion along each eigencomponent to identify task-variant directions and applies adaptive soft-shrinkage to suppress task-induced noise while preserving task-invariant semantic structure, without inference-time overhead. Experiments across multiple embedding backbones and specialized benchmarks show that REZE outperforms standard pre-finetuning and isotropy-oriented post-hoc regularization in most settings, remaining stable where existing PFT variants collapse. Embedding space analyses further confirm that REZE induces controlled shifts aligned with the original embedding manifold, underscoring representation shift control as a key principle for robust embedding pre-finetuning under heterogeneous supervision.

cs.CL↗

arXiv2026

Estimating Long Run Welfare Outcome in Rotating Panel with Grouped Fixed Effects: Application to Poverty Dynamics in Peru

Household welfare dynamics are often difficult to investigate due to lack of long-term panel data. Existing methods, such as pseudo-panel and synthetic panel, offer widely used solutions based on repeated cross-section designs, but they do not exploit within-household variation in rotating panel designs, which provide very useful information for estimating long-run dynamics. This paper applies grouped fixed effects (GFE) to estimate poverty mobility and persistence in a rotating panel setting, using National Household Survey on Living Conditions and Poverty (ENAHO) in Peru. Using observed transitions, we show that GFE-implied poverty transitions closely track the data. In a one-step-ahead validation that holds out each household's final observed year, predicted transition shares remain close to realized transition shares, indicating that the method captures short-run entry and exit dynamics out of sample. When benchmarked against synthetic panel point estimates, the GFE approach delivers transition measures that are closer to observed transitions on average, while also providing an interpretable grouping structure that supports richer descriptions of poverty persistence and mobility.

econ.EM↗

arXiv2026

Let Triggers Control: Frequency-Aware Dropout for Effective Token Control

Text-to-image models such as Stable Diffusion have achieved unprecedented levels of high-fidelity visual synthesis. As these models advance, personalization of generative models -- commonly facilitated through Low-Rank Adaptation (LoRA) with a dedicated trigger token -- has become a significant area of research. Previous works have naively assumed that fine-tuning with a single trigger token to represent new concepts. However, this often results in poor controllability, where the trigger token alone fails to reliably evoke the intended concept. We attribute this issue to the frequent co-occurrence of the trigger token with the surrounding context during fine-tuning, which entangles their representations and compromises the token's semantic distinctiveness. To disentangle this, we propose Frequency-Aware Dropout (FAD) -- a novel regularization technique that improves prompt controllability without adding new parameters. FAD consists of two key components: co-occurrence analysis and curriculum-inspired scheduling. Qualitative and quantitative analyses across token-based diffusion models (SD~1.5 and SDXL) and natural language--driven backbones (FLUX and Qwen-Image) demonstrate consistent gains in prompt fidelity, stylistic precision, and user-perceived quality. Our method provides a simple yet effective dropout strategy that enhances controllability and personalization in text-to-image generation. Notably, it achieves these improvements without introducing additional parameters or architectural modifications, making it readily applicable to existing models with minimal computational overhead.

cs.CV↗

arXiv2026

Automatic Inter-document Multi-hop Scientific QA Generation

Existing automatic scientific question generation studies mainly focus on single-document factoid QA, overlooking the inter-document reasoning crucial for scientific understanding. We present AIM-SciQA, an automated framework for generating multi-document, multi-hop scientific QA datasets. AIM-SciQA extracts single-hop QAs using large language models (LLMs) with machine reading comprehension and constructs cross-document relations based on embedding-based semantic alignment while selectively leveraging citation information. Applied to 8,211 PubMed Central papers, it produced 411,409 single-hop and 13,672 multi-hop QAs, forming the IM-SciQA dataset. Human and automatic validation confirmed high factual consistency, and experimental results demonstrate that IM-SciQA effectively differentiates reasoning capabilities across retrieval and QA stages, providing a realistic and interpretable benchmark for retrieval-augmented scientific reasoning. We further extend this framework to construct CIM-SciQA, a citation-guided variant achieving comparable performance to the Oracle setting, reinforcing the dataset's validity and generality.

cs.CL↗

arXiv2025

Ultrafast exciton polaron dynamics in 2D Ruddlesden Popper lead halide perovskites

Two dimensional Ruddlesden Popper (2D) RP hybrid perovskites exhibit substantially higher chemical and structural stability than their three dimensional (3D) counterparts, positioning them as promising candidates for next generation optoelectronics. While quasiparticle dynamics in 3D perovskites are well studied, their 2D analogues remain comparatively underexplored. Here we systematically investigate the branching, dynamics, and interactions of free excitons (FEs) and exciton polarons EPs in monolayer 2D RP perovskites using visible range femtosecond transient absorption TA spectroscopy. We prepared monolayer 2D RP perovskite thin films with varied organic spacers and distinct fabrication routes for comparative analysis. We find that the EP binding energy is 50 65 meV in (BA)2PbI4 and 37 39 meV in (PEA)2PbI4, consistent with spacer layer dependent coupling as corroborated by FTIR. We reveal a dynamic equilibrium between FEs and EPs that persists for tens of picoseconds. Notably, the TA signatures differ by fabrication route films from the newly developed process show weaker Auger annihilation and a reduced hot phonon bottleneck than those from the conventional route trends consistent with fewer traps and impurities in the former. Coupled rate equation modeling reproduces the transients and quantifies the processes of hot carrier relaxation, exciton exciton annihilation, exciton phonon coupling, and FE EP interconversion. These results demonstrate that the chemical synthetic process (fabrication route) and spacer choice significantly influence EP stability and population balance, offering practical levers for engineering ultrafast photophysics in 2D perovskites and guiding the design of advanced optoelectronic devices.

cond-mat.mtrl-sci↗

arXiv2025

LTGS: Long-Term Gaussian Scene Chronology From Sparse View Updates

Recent advances in novel-view synthesis can create the photo-realistic visualization of real-world environments from conventional camera captures. However, the everyday environment experiences frequent scene changes, which require dense observations, both spatially and temporally, that an ordinary setup cannot cover. We propose long-term Gaussian scene chronology from sparse-view updates, coined LTGS, an efficient scene representation that can embrace everyday changes from highly under-constrained casual captures. Given an incomplete and unstructured 3D Gaussian Splatting (3DGS) representation obtained from an initial set of input images, we robustly model the long-term chronology of the scene despite abrupt movements and subtle environmental variations. We construct objects as template Gaussians, which serve as structural, reusable priors for shared object tracks. Then, the object templates undergo a further refinement pipeline that modulates the priors to adapt to temporally varying environments given few-shot observations. Once trained, our framework is generalizable across multiple time steps through simple transformations, significantly enhancing the scalability for a temporal evolution of 3D environments. As existing datasets do not explicitly represent the long-term real-world changes with a sparse capture setup, we collect real-world datasets to evaluate the practicality of our pipeline. Experiments demonstrate that our framework achieves superior reconstruction quality compared to other baselines while enabling fast and light-weight updates. Project page is available at: https://mkjjang3598.github.io/LTGS.

cs.CV↗

arXiv2025

The Probability of Food Security: A new longitudinal data set using the Panel Study of Income Dynamics

The study of food security dynamics in the U.S. has long been impeded by the lack of extended longitudinal observations of the same households or individuals. This paper applies a newly-introduced household-level food security measure, the probability of food security (PFS), to 26 waves of Panel Study of Income Dynamics (PSID) data, spanning 1979-2019, to generate a data product we describe and make newly available to the research community. We detail the construction of this unprecedentedly long food security panel data series in PSID data. Finally, we estimate key subpopulation- and national-level food security dynamics identifiable over the 40-year (1979-2019) period spanning multiple recessions and federal nutrition assistance policy changes, including disaggregated dynamics based on geography, race, sex, and educational attainment.

econ.GN↗

arXiv2025

Performance Analysis of IEEE 802.11bn with Coordinated TDMA on Real-Time Applications

Wi-Fi plays a crucial role in connecting electronic devices and providing communication services in everyday life. Recently, there has been a growing demand for services that require low-latency communication, such as real-time applications. The latest amendments to Wi-Fi, IEEE 802.11bn, are being developed to address these demands with technologies such as the multiple access point coordination (MAPC). In this paper, we demonstrate that coordinated TDMA (Co-TDMA), one of the MAPC techniques, effectively reduces the latency of transmitting time-sensitive traffic. In particular, we focus on worst-case latency and jitter, which are key metrics for evaluating the performance of real-time applications. We first introduce a Co-TDMA scheduling strategy. We then investigate how this scheduling strategy impacts latency under varying levels of network congestion and traffic volume characteristics. Finally, we validate our findings through system-level simulations. Our simulation results demonstrate that Co-TDMA effectively mitigates jitter and worst-case latency for low-latency traffic, with the latter exhibiting an improvement of approximately 24%.

cs.IT↗

arXiv2025

How much does SNAP Matter? SNAP's Effects on Food Security

Supplemental Nutrition Assistance Program (SNAP) aims to improve food security of low-income households in the U.S. A new, continuous food security measure called the Probability of Food Security (PFS), which proxies for the official food security measure but is implementable on longer periods, enables the study of SNAP's effects on the intensive margin. Using variations in state-level SNAP administrative policies as an instrument for individual SNAP participation, I find that SNAP does not have significant effects on estimated food security on average, both on the entire population and low-income population whom I defined as income is below 130\% of poverty line at least once during the study period. I find SNAP has stronger positive effects on those whose estimated food security status is in the middle of the distribution, but has no significant effects in the tails of the distribution.

econ.GN↗

arXiv2025

Event-Driven Storytelling with Multiple Lifelike Humans in a 3D Scene

In this work, we propose a framework that creates a lively virtual dynamic scene with contextual motions of multiple humans. Generating multi-human contextual motion requires holistic reasoning over dynamic relationships among human-human and human-scene interactions. We adapt the power of a large language model (LLM) to digest the contextual complexity within textual input and convert the task into tangible subproblems such that we can generate multi-agent behavior beyond the scale that was not considered before. Specifically, our event generator formulates the temporal progression of a dynamic scene into a sequence of small events. Each event calls for a well-defined motion involving relevant characters and objects. Next, we synthesize the motions of characters at positions sampled based on spatial guidance. We employ a high-level module to deliver scalable yet comprehensive context, translating events into relative descriptions that enable the retrieval of precise coordinates. As the first to address this problem at scale and with diversity, we offer a benchmark to assess diverse aspects of contextual reasoning. Benchmark results and user studies show that our framework effectively captures scene context with high scalability. The code and benchmark, along with result videos, are available at our project page: https://rms0329.github.io/Event-Driven-Storytelling/.

cs.CV↗

arXiv2025

Def-DTS: Deductive Reasoning for Open-domain Dialogue Topic Segmentation

Dialogue Topic Segmentation (DTS) aims to divide dialogues into coherent segments. DTS plays a crucial role in various NLP downstream tasks, but suffers from chronic problems: data shortage, labeling ambiguity, and incremental complexity of recently proposed solutions. On the other hand, Despite advances in Large Language Models (LLMs) and reasoning strategies, these have rarely been applied to DTS. This paper introduces Def-DTS: Deductive Reasoning for Open-domain Dialogue Topic Segmentation, which utilizes LLM-based multi-step deductive reasoning to enhance DTS performance and enable case study using intermediate result. Our method employs a structured prompting approach for bidirectional context summarization, utterance intent classification, and deductive topic shift detection. In the intent classification process, we propose the generalizable intent list for domain-agnostic dialogue intent classification. Experiments in various dialogue settings demonstrate that Def-DTS consistently outperforms traditional and state-of-the-art approaches, with each subtask contributing to improved performance, particularly in reducing type 2 error. We also explore the potential for autolabeling, emphasizing the importance of LLM reasoning techniques in DTS.

cs.CL↗

arXiv2024

Innovation beyond intention: harnessing exaptation for technological breakthroughs

The frameworks that explore scientific and technological evolution suggest that discoveries and inventions are intrinsic processes, while the wealth of knowledge accumulated over time enables researchers to make further advancements, echoing Newton's sentiment of "standing on the shoulders of giants." Despite the exponential growth in new scientific and technical knowledge, the consolidation-disruption (D) index suggests a concerning decline in the disruptiveness of papers and patents. "Exaptation" a concept borrowed from biological evolution, is now recognized as a pivotal yet often neglected mechanism in technological evolution. Significant technologies often do not emerge out of thin air but rather result from the application of existing technologies in other domains. For instance, bird feathers initially served as waterproofing and insulation before enabling flight, and microwave ovens originated from radar magnetrons. Exaptation, acknowledged as the catalyst for "innovation beyond intention" signifies a cross-field evolutionary process that is driven by functional shifts in pre-existing knowledge, technology, or artifacts. In this study, we introduce the concept of exaptation value, deliberately excluding serendipity. Our analysis reveals that, despite a declining trend in the disruptiveness of innovation, there is an increasing trend in the application of cross-domain knowledge within the innovation process over time. We also explore the impact of technology exaptation on innovation disruptiveness and discuss how leveraging technology adaptability enhances innovation's disruptive potential.

physics.soc-ph↗

arXiv2024

Context-Aware LLM Translation System Using Conversation Summarization and Dialogue History

Translating conversational text, particularly in customer support contexts, presents unique challenges due to its informal and unstructured nature. We propose a context-aware LLM translation system that leverages conversation summarization and dialogue history to enhance translation quality for the English-Korean language pair. Our approach incorporates the two most recent dialogues as raw data and a summary of earlier conversations to manage context length effectively. We demonstrate that this method significantly improves translation accuracy, maintaining coherence and consistency across conversations. This system offers a practical solution for customer support translation tasks, addressing the complexities of conversational text.

cs.CL↗

arXiv2024

Very-Large-Scale GPU-Accelerated Nuclear Gradient of Time-Dependent Density Functional Theory with Tamm-Dancoff Approximation and Range-Separated Hybrid Functionals

Modern graphics processing units (GPUs) provide an unprecedented level of computing power. In this study, we present a high-performance, multi-GPU implementation of the analytical nuclear gradient for Kohn-Sham time-dependent density functional theory (TDDFT), employing the Tamm-Dancoff approximation (TDA) and Gaussian-type atomic orbitals as basis functions. We discuss GPU-efficient algorithms for the derivatives of electron repulsion integrals and exchange-correlation functionals within the range-separated scheme. As an illustrative example, we calculated the TDA-TDDFT gradient of the S1 state of a full-scale green fluorescent protein with explicit water solvent molecules, totaling 4353 atoms, at the wB97X/def2-SVP level of theory. Our algorithm demonstrates favorable parallel efficiencies on a high-speed distributed system equipped with 256 Nvidia A100 GPUs, achieving >70% with up to 64 GPUs and 31% with 256 GPUs, effectively leveraging the capabilities of modern high-performance computing systems.

physics.chem-ph↗

arXiv2020

HetPipe: Enabling Large DNN Training on (Whimpy) Heterogeneous GPU Clusters through Integration of Pipelined Model Parallelism and Data Parallelism

Deep Neural Network (DNN) models have continuously been growing in size in order to improve the accuracy and quality of the models. Moreover, for training of large DNN models, the use of heterogeneous GPUs is inevitable due to the short release cycle of new GPU architectures. In this paper, we investigate how to enable training of large DNN models on a heterogeneous GPU cluster that possibly includes whimpy GPUs that, as a standalone, could not be used for training. We present a DNN training system, HetPipe (Heterogeneous Pipeline), that integrates pipelined model parallelism (PMP) with data parallelism (DP). In HetPipe, a group of multiple GPUs, called a virtual worker, processes minibatches in a pipelined manner, and multiple such virtual workers employ data parallelism for higher performance. We also propose a novel parameter synchronization model, which we refer to as Wave Synchronous Parallel (WSP) to accommodate both PMP and DP for virtual workers, and provide convergence proof of WSP. Our experimental results on a given heterogeneous setting show that with HetPipe, DNN models converge up to 49% faster compared to the state-of-the-art DP technique.

cs.DC↗