arXiv subjects

Jiwon Seo

Publications and source records attributed to Jiwon Seo.

At least 19 recordsLinked to original sources

arXiv2026

Quantum-Based k-Coverage Optimization for UAV-Aided Search and Rescue Missions

In large-scale disaster scenarios, rapid localization of missing persons is a critical challenge for search-and-rescue (SAR) operations. Unmanned aerial vehicles (UAVs) equipped with radio frequency (RF) receivers can support RF-based localization by collecting signals emitted from mobile devices at spatially distributed sensing locations. This paper addresses the resulting waypoint-selection problem: determining a minimum set of UAV waypoints that provides at least threefold coverage of every potential target location. We formulate this task as an extended k-coverage problem that independently defines the UAV-navigable and target regions, and derive an exact-penalty quadratic unconstrained binary optimization (QUBO) formulation with a sufficient penalty condition that preserves feasibility and minimum waypoint cardinality. The QUBO is mapped to an Ising-form cost Hamiltonian and evaluated using the quantum approximate optimization algorithm (QAOA) on both a noise-free simulator and IBM's 127-qubit Eagle processor. On the tested simulator instances, QAOA recovers the known minimum-cardinality solutions. Across the rectangular hardware test cases, the mean 3-coverage ratio exceeded 95%. In the campus-scale evaluation, ten hardware executions achieved 99.3% mean 3-coverage with a 90% feasible-run rate, while the shortest feasible flight path was up to 37.0% shorter than those of the deterministic grid-based baselines. Additional comparisons with classical optimization and learning-based baselines are provided, together with computational and quantum-resource analyses for larger generated instances. These results establish an exact QUBO representation for RF-based SAR waypoint selection and characterize its implementation on current gate-based quantum hardware.

eess.SY↗

arXiv2026

Asymmetric Collapse in Model Merging: When Refusal Over- writes Recognition

Model merging is often used to combine capabilities from separately fine-tuned models without additional training, but it is unclear whether standard merging methods preserve multiple safety-relevant behaviors simultaneously. We study this question through a controlled case study using two Gemma-3-1B-IT finetunes on two complementary safety objectives: CARES harm-level classification and WildJailbreak adversarial refusal. We merge the two fine-tunes using Linear, SLERP, TIES, and DARE-TIES, and evaluate the merged models on classification accuracy, attack resistance, and benign compliance. Across all four methods, attack resistance transfers significantly more than classification accuracy: merged models retain 81-85% jailbreak refusal rates while CARES accuracy falls to at most 12.9%. Weight-space measurements suggest that this asymmetry is not caused by strongly opposing task-vector directions: the two task vectors are nearly orthogonal (cosine similarity 0.011). Instead, the refusal fine-tune induces consistently larger per-layer task-vector magnitudes, causing magnitude-sensitive methods to favor refusal updates. These results show that standard model merging can collapse safety recognition into broad refusal when safety-relevant task vectors differ substantially in scale.

cs.AI↗

arXiv2026

Rigidity-Based Multi-UAV Trajectory Optimization for Rapid Cooperative Emergency Target Localization

Reducing the response time for accurate emergency-caller localization is critical in vehicular and public-safety networks. Although mobile devices commonly use GNSS, Wi-Fi, or cellular positioning, their accuracy and availability can degrade because of poor signal reception, limited infrastructure, and regulatory constraints. UAV-based localization offers a promising alternative by using airborne sensors to cooperatively estimate the target position. However, existing Fisher information matrix (FIM)-based trajectory optimization methods depend on the current target estimate and can perform poorly in the early mission stage, when measurements are limited and uncertainty is high. We propose a rigidity-based UAV trajectory optimization method that maximizes the smallest nonzero singular value of the rigidity matrix associated with the UAV-target sensing graph, improving geometric conditioning and reducing position ambiguity. We also introduce a pruning-based matrix reduction strategy for efficient real-time implementation. Simulations show that the proposed method reduces search time by 32.9% compared with FIM-based methods and satisfies the FCC horizontal emergency-localization requirement sooner. Further results demonstrate scalability, robustness to UAV positioning errors and NLOS path loss, low sensitivity to heading parameters, practical computation and communication costs, and more stable degradation than PPO-based baselines under severe sensing and navigation perturbations.

eess.SY↗

arXiv2026

Robust Ultra Low-Bit Post-Training Quantization via Stable Diagonal Curvature Estimate

Large Language Models (LLMs) are widely used across many domains, but their scale makes deployment challenging. Post-Training Quantization (PTQ) reduces memory footprint without retraining by leveraging a small calibration set. Recent Hessian-based PTQ methods compensate quantization error via cross-channel dependencies, but such approaches degrade at low bit-widths due to noisy curvature estimates from limited calibration data. We propose DASH-Q, a robust PTQ framework using diagonal Hessian approximation and iterative weighted least squares. By discarding noise-prone dependencies, DASH-Q filters sampling noise while prioritizing the preservation of salient feature power. We outperform other PTQ baselines in ultra low-bit regime, improving zero-shot accuracy by 7.01% on average and up to 14.01% over the strongest baselines across five baseline LLM models, while showing robust and stable performance with very small calibration data.

cs.LG↗

arXiv2026

Zonotope Shadow and Reflection Matching: A Novel GNSS Reflection-Based Framework for Enhanced Positioning Accuracy in Urban Areas

In urban areas, signal reception conditions are often poor due to reflections from buildings, resulting in inaccurate global navigation satellite system (GNSS)-based positioning. Various 3D-mapping-aided (3DMA) GNSS techniques, including shadow matching, have been proposed to address this issue. However, conventional shadow matching estimates positions in a discretized manner. The accuracy of this approach is limited by the resolution of the grid points representing the candidate receiver positions, making it difficult to achieve robust urban positioning and to ensure that the position estimate satisfies user-specified protection levels or safety bounds. To overcome these limitations, zonotope shadow matching (ZSM) has been proposed, which utilizes a set-based position estimate rather than grid-based estimates. ZSM calculates the GNSS shadow--an area on the ground where the line-of-sight (LOS) is blocked and only non-line-of-sight (NLOS) signals can be received--to estimate the receiver's position set. ZSM distinguishes between LOS and NLOS satellites, determining that the receiver is inside the GNSS shadow if the satellite is NLOS and outside if the satellite is LOS. However, relying solely on GNSS shadows limits the ability to sufficiently reduce the size of the receiver position set and to precisely estimate the receiver's location. To address this, we propose zonotope shadow and reflection matching (ZSRM) to enhance positioning accuracy in urban areas. The proposed ZSRM technique is validated through field tests using GNSS signals collected in an urban environment. Consequently, the RMS horizontal position error of ZSRM improved by 10.0% to 53.6% compared with ZSM, while the RMS cross-street and along-street position bounds improved by 18.0% to 50.1% and 30.7% to 59.3%, respectively.

eess.SP↗

arXiv2025

FlexiQ: Adaptive Mixed-Precision Quantization for Latency/Accuracy Trade-Offs in Deep Neural Networks

Neural networks commonly execute on hardware accelerators such as NPUs and GPUs for their size and computation overhead. These accelerators are costly and it is hard to scale their resources to handle real-time workload fluctuations. We present FlexiQ, an adaptive mixed-precision quantization scheme for computer vision models. FlexiQ selectively applies low-bitwidth computation to feature channels with small value ranges and employs an efficient bit-lowering method to minimize quantization errors while maintaining inference accuracy. Furthermore, FlexiQ adjusts its low-bitwidth channel ratio in real time, enabling quantized models to effectively manage fluctuating inference workload. We implemented FlexiQ prototype, including the mixed-precision inference runtime on our custom NPU and GPUs. Evaluated on eleven convolution- and transformer-based vision models, FlexiQ achieves on average 6.6% higher accuracy for 4-bit models with finetuning and outperforms four state-of-the-art quantization techniques. Moreover, our mixed-precision models achieved an efficient accuracy-latency trade-off, with the 50% 4-bit model incurring only 0.6% accuracy loss while achieving 40% of the speedup of the 100% 4-bit model over 8-bit model. Latency evaluations on our NPU and GPUs confirmed that FlexiQ introduces minimal runtime overhead, demonstrating its hardware efficiency and overall performance benefits.

cs.LG↗

arXiv2025

Speculative Verification: Exploiting Information Gain to Refine Speculative Decoding

LLMs have low GPU efficiency and high latency due to autoregressive decoding. Speculative decoding (SD) mitigates this using a small draft model to speculatively generate multiple tokens, which are then verified in parallel by a target model. However, when speculation accuracy is low, the overhead from rejected tokens can offset the benefits, limiting SD's effectiveness, especially at large batch sizes. To address this, we propose Speculative Verification (SV), an efficient augmentation to SD that dynamically predicts speculation accuracy and adapts the verification length to maximize throughput. SV introduces a companion model - a small auxiliary model similar in size to the draft model - to estimate the alignment between draft and target model distributions. By maximizing the information gain from quantifying this alignment, SV refines verification decisions, reducing wasted computation on rejected tokens and improving decoding efficiency. Moreover, SV requires no modifications to the draft or target models and is compatible with existing SD variants. We extensively evaluated SV on publicly available LLMs across three NLP tasks using nine combinations of draft, companion, and target models, including 13B-72B target models and three types of variations: base (no finetuning), instruction-tuned, and task fine-tuned. Across all experiments and batch sizes (4-80), SV consistently outperforms both SD and standard decoding with the target model. It improves SD performance by up to 2$\times$, with an average speedup of 1.4 $\times$ in large-batch settings (batch sizes 32-80). These results demonstrate SV's robustness, scalability, and practical utility for efficient LLM inference.

cs.CL↗

arXiv2025

Enhancing Urban GNSS Positioning Reliability via Conservative Satellite Selection Using Unanimous Voting Across Multiple Machine Learning Classifiers

In urban environments, global navigation satellite system (GNSS) positioning is often compromised by signal blockages and multipath effects caused by buildings, leading to significant positioning errors. To address this issue, this study proposes a robust enhancement of zonotope shadow matching (ZSM)-based positioning by employing a conservative satellite selection strategy using unanimous voting across multiple machine learning classifiers. Three distinct models - random forest (RF), gradient boosting decision tree (GBDT), and support vector machine (SVM) - were trained to perform line-of-sight (LOS) and non-line-of-sight (NLOS) classification based on global positioning system (GPS) signal features. A satellite is selected for positioning only when all classifiers unanimously agree on its classification and their associated confidence scores exceed a threshold. Experiments with real-world GPS data collected in dense urban areas demonstrate that the proposed method significantly improves the positioning success rate and the receiver containment rate, even with imperfect LOS/NLOS classification. Although a slight increase in the position bound was observed due to the reduced number of satellites used, overall positioning reliability was substantially enhanced, indicating the effectiveness of the proposed approach in urban GNSS environments.

eess.SP↗

arXiv2025

Enhancing eLoran Timing Accuracy via Machine Learning with Meteorological and Terrain Data

The vulnerabilities of global navigation satellite systems (GNSS) to signal interference have increased the demand for complementary positioning, navigation, and timing (PNT) systems. To address this, South Korea has decided to deploy an enhanced long-range navigation (eLoran) system as a complementary PNT solution. Similar to GNSS, eLoran provides highly accurate timing information, which is essential for applications such as telecommunications, financial systems, and power distribution. However, the primary sources of error for GNSS and eLoran differ. For eLoran, the main source of error is signal propagation delay over land, known as the additional secondary factor (ASF). This delay, influenced by ground conductivity and weather conditions along the signal path, is challenging to predict and mitigate. In this paper, we measure the time difference (TD) between GPS and eLoran using a time interval counter and analyze the correlations between eLoran/GPS TD and eleven meteorological factors. Accurate estimation of eLoran/GPS TD could enable eLoran to achieve timing accuracy comparable to that of GPS. We propose two estimation models for eLoran/GPS TD and compare their performance with existing TD estimation methods. The proposed WLR-AGRNN model captures the linear relationships between meteorological factors and eLoran/GPS TD using weighted linear regression (WLR) and models nonlinear relationships between outputs from expert networks through an anisotropic general regression neural network (AGRNN). The model incorporates terrain elevation to appropriately weight meteorological data, as elevation influences signal propagation delay. Experimental results based on four months of data demonstrate that the WLR-AGRNN model outperforms other models, highlighting its effectiveness in improving eLoran/GPS TD estimation accuracy.

eess.SP↗

arXiv2025

Toward High Accuracy DME for Alternative Aircraft Positioning: SFOL Pulse Transmission in High-Power DME

The Stretched-FrOnt-Leg (SFOL) pulse is an advanced distance measuring equipment (DME) pulse that offers superior ranging accuracy compared to conventional Gaussian pulses. Successful SFOL pulse transmission has been recently demonstrated from a commercial Gaussian pulse-based DME in low-power mode utilizing digital predistortion (DPD) techniques for power amplifiers. These adjustments were achieved through software modifications, enabling SFOL integration without replacing existing DME infrastructure. However, the SFOL pulse is designed to optimize ranging capabilities by leveraging the effective radiated power (ERP) and pulse shape parameters permitted within DME specifications. Consequently, it operates with narrow margins against these specifications, potentially leading to non-compliance when transmitted in high-power mode. This paper introduces strategies to enable a Gaussian pulse-based DME to transmit the SFOL pulse while adhering to DME specifications in high-power mode. The proposed strategies involve use of a variant of the SFOL pulse and DPD techniques utilizing truncated singular value decomposition, tailored for high-power DME operations. Test results, conducted on a testbed utilizing a commercial Gaussian pulse-based DME, demonstrate the effectiveness of these strategies, ensuring compliance with DME specifications in high-power mode with minimal performance loss. This study enables cost-effective integration of high-accuracy SFOL pulses into existing high-power DME systems, enhancing aircraft positioning precision while ensuring compliance with industry standards.

eess.SP↗

arXiv2025

Set-Based Position Ambiguity Reduction Method for Zonotope Shadow Matching in Urban Areas Using Estimated Multipath Errors

In urban areas, the quality of global navigation satellite system (GNSS) signals deteriorates, leading to reduced positioning accuracy. To address this issue, 3D-mapping-aided (3DMA) techniques, such as shadow matching and zonotope shadow matching (ZSM), have been proposed. However, these methods can introduce a problem known as multi-modal position ambiguity, making it challenging to select the exact mode in which the receiver is located. Accurately selecting the correct mode is essential for improving positioning accuracy. A previous study proposed a method that uses satellite-pseudorange consistency (SPC), calculated from pseudorange measurements, to select the mode containing the receiver. This method achieved a mode selection accuracy of approximately 78%. To further enhance accuracy, the study utilized pseudorange measurements collected at multiple timesteps from a fixed location and a trained line-of-sight (LOS) classifier. However, in practice, collecting data at multiple timesteps from the same location in dynamic environments is challenging. Moreover, the performance of the trained LOS classifier heavily depends on the surrounding environment, leading to low reliability. In this study, we propose a method that estimates and corrects multipath errors based on the mode distribution obtained from the output of ZSM and extract an enhanced SPC using the corrected pseudorange measurements. This enables high mode selection accuracy using only single-timestep pseudorange measurements, without requiring a trained LOS classifier.

eess.SP↗

arXiv2025

Reducing Computational Complexity of Rigidity-Based UAV Trajectory Optimization for Real-Time Cooperative Target Localization

Accurate and swift localization of the target is crucial in emergencies. However, accurate position data of a target mobile device, typically obtained from global navigation satellite systems (GNSS), cellular networks, or WiFi, may not always be accessible to first responders. For instance, 1) accuracy and availability can be limited in challenging signal reception environments, and 2) in regions where emergency location services are not mandatory, certain mobile devices may not transmit their location during emergencies. As an alternative localization method, a network of unmanned aerial vehicles (UAVs) can be employed to passively locate targets by collecting radio frequency (RF) signal measurements, such as received signal strength (RSS). In these situations, UAV trajectories play a critical role in localization performance, influencing both accuracy and search time. Previous studies optimized UAV trajectories using the determinant of the Fisher information matrix (FIM), but its performance declines under unfavorable geometric conditions, such as when UAVs start from a single base, leading to position ambiguity. To address this, our prior work introduced a rigidity-based approach, which improved the search time compared to FIM-based methods in our simulation case. However, the high computational cost of rigidity-based optimization, primarily due to singular value decomposition (SVD), limits its practicality. In this paper, we applied techniques to reduce computational complexity, including randomized SVD, smooth SVD, and vertex pruning.

eess.SY↗

arXiv2024

Quantum Support Vector Machine-Based Classification of GPS Signal Reception Conditions

Global Positioning System (GPS) plays a critical role in navigation by utilizing satellite signals, but its accuracy in urban environments is often compromised by signal obstructions. Previous research has categorized GPS reception conditions into line-of-sight (LOS), non-line-of-sight (NLOS), and LOS+NLOS scenarios to enhance accuracy. This paper introduces a novel approach using quantum support vector machines (QSVM) with a ZZ feature map and fidelity quantum kernel to classify urban GPS signal reception conditions, comparing its performance against classical SVM methods. While classical SVM has been previously explored for this purpose, our study is the first to apply QSVM to this classification task. We conducted experiments using datasets from two distinct urban locations to train and evaluate SVM and QSVM models. Our results demonstrate that QSVM achieves superior classification accuracy compared to classical SVM for urban GPS signal datasets. Additionally, we emphasize the importance of appropriately scaling raw data when utilizing QSVM.

eess.SP↗

arXiv2024

HELPS for Emergency Location Service: Hyper-Enhanced Local Positioning System

In this study, we propose a novel positioning and searching system for emergency location services, namely the hyper-enhanced local positioning system (HELPS), which is applicable to all mobile phone users, including legacy feature phone users. In the case of an emergency, rescuers are dispatched with portable signal measurement equipment around the estimated location of the emergency caller. Each signal measurement device measures the uplink signal from the mobile phone of the caller. After calculating the rough location of the caller's mobile phone based on these measurements, rescuers can efficiently search for the caller using the received uplink signal strength. Thus, the positioning accuracy in a conventional sense is not a limitation for rescuers in finding the caller. HELPS is not a traditional positioning system but rather a system with humans in the loop designed to reduce search time in emergencies. HELPS can provide emergency location information even in environments where the GPS or Wi-Fi is not functional. Furthermore, for HELPS operation, no hardware changes or software installations are required on the caller's mobile phone.

eess.SP↗

arXiv2024

Benchmarks Underestimate the Readiness of Multi-lingual Dialogue Agents

Creating multilingual task-oriented dialogue (TOD) agents is challenging due to the high cost of training data acquisition. Following the research trend of improving training data efficiency, we show for the first time, that in-context learning is sufficient to tackle multilingual TOD. To handle the challenging dialogue state tracking (DST) subtask, we break it down to simpler steps that are more compatible with in-context learning where only a handful of few-shot examples are used. We test our approach on the multilingual TOD dataset X-RiSAWOZ, which has 12 domains in Chinese, English, French, Korean, Hindi, and code-mixed Hindi-English. Our turn-by-turn DST accuracy on the 6 languages range from 55.6% to 80.3%, seemingly worse than the SOTA results from fine-tuned models that achieve from 60.7% to 82.8%; our BLEU scores in the response generation (RG) subtask are also significantly lower than SOTA. However, after manual evaluation of the validation set, we find that by correcting gold label errors and improving dataset annotation schema, GPT-4 with our prompts can achieve (1) 89.6%-96.8% accuracy in DST, and (2) more than 99% correct response generation across different languages. This leads us to conclude that current automatic metrics heavily underestimate the effectiveness of in-context learning.

cs.CL↗

arXiv2024

Efficient Quantum Circuit Encoding of Object Information in 2D Ray Casting

Quantum computing holds the potential to solve problems that are practically unsolvable by classical computers due to its ability to significantly reduce time complexity. We aim to harness this potential to enhance ray casting, a pivotal technique in computer graphics for simplifying the rendering of 3D objects. To perform ray casting in a quantum computer, we need to encode the defining parameters of primitives into qubits. However, during the current noisy intermediate-scale quantum (NISQ) era, challenges arise from the limited number of qubits and the impact of noise when executing multiple gates. Through logic optimization, we reduced the depth of quantum circuits as well as the number of gates and qubits. As a result, the event count of correct measurements from an IBM quantum computer significantly exceeded that of incorrect measurements.

quant-ph↗

arXiv2024

ExeGPT: Constraint-Aware Resource Scheduling for LLM Inference

This paper presents ExeGPT, a distributed system designed for constraint-aware LLM inference. ExeGPT finds and runs with an optimal execution schedule to maximize inference throughput while satisfying a given latency constraint. By leveraging the distribution of input and output sequences, it effectively allocates resources and determines optimal execution configurations, including batch sizes and partial tensor parallelism. We also introduce two scheduling strategies based on Round-Robin Allocation and Workload-Aware Allocation policies, suitable for different NLP workloads. We evaluate ExeGPT on six LLM instances of T5, OPT, and GPT-3 and five NLP tasks, each with four distinct latency constraints. Compared to FasterTransformer, ExeGPT achieves up to 15.2x improvements in throughput and 6x improvements in latency. Overall, ExeGPT achieves an average throughput gain of 2.9x across twenty evaluation scenarios. Moreover, when adapting to changing sequence distributions, the cost of adjusting the schedule in ExeGPT is reasonably modest. ExeGPT proves to be an effective solution for optimizing and executing LLM inference for diverse NLP workload and serving conditions.

cs.DC↗

arXiv2023

IR-UWB Radar-Based Contactless Silent Speech Recognition of Vowels, Consonants, Words, and Phrases

Several sensing techniques have been proposed for silent speech recognition (SSR); however, many of these methods require invasive processes or sensor attachment to the skin using adhesive tape or glue, rendering them unsuitable for frequent use in daily life. By contrast, impulse radio ultra-wideband (IR-UWB) radar can operate without physical contact with users' articulators and related body parts, offering several advantages for SSR. These advantages include high range resolution, high penetrability, low power consumption, robustness to external light or sound interference, and the ability to be embedded in space-constrained handheld devices. This study demonstrated IR-UWB radar-based contactless SSR using four types of speech stimuli (vowels, consonants, words, and phrases). To achieve this, a novel speech feature extraction algorithm specifically designed for IR-UWB radar-based SSR is proposed. Each speech stimulus is recognized by applying a classification algorithm to the extracted speech features. Two different algorithms, multidimensional dynamic time warping (MD-DTW) and deep neural network-hidden Markov model (DNN-HMM), were compared for the classification task. Additionally, a favorable radar antenna position, either in front of the user's lips or below the user's chin, was determined to achieve higher recognition accuracy. Experimental results demonstrated the efficacy of the proposed speech feature extraction algorithm combined with DNN-HMM for classifying vowels, consonants, words, and phrases. Notably, this study represents the first demonstration of phoneme-level SSR using contactless radar.

eess.AS↗