arXiv subjects

Seungwoo Jeong

Publications and source records attributed to Seungwoo Jeong.

11 recordsLinked to original sources

arXiv2026

Latency Prediction for LLM Inference on NPU Systems

Deploying Large Language Models (LLMs) requires exploring a large configuration space spanning parallelization strategies, batching techniques, and scheduling policies. Exhaustive measurement across this space is impractical, making latency prediction essential for system optimization. While NPUs have emerged as accelerators designed for LLM inference, no prediction methodology has been established for them. Specifically, applying prior work to LLM inference latency prediction on NPUs faces three challenges: undisclosed microarchitecture of commercial NPUs, unpredictable compiler optimizations, and latency non-linearity induced by bucketing. We present LENS, a latency estimator that predicts NPU inference latency without information on the microarchitecture or compiler, and captures the non-linear latency induced by bucketing. LENS profiles each bucket with two end-to-end (E2E) measurements and composes the results to predict latency for arbitrary input-output length combinations. We validate LENS across NPUs from multiple vendors, several LLMs, and diverse workloads, achieving a mean prediction error of 2.15\%. We further compare LENS against two methodologically related baselines, confirming the validity of its approach.

cs.DC↗

arXiv2026

ShuntServe: Cost-Efficient LLM Serving on Heterogeneous Spot GPU Clusters

As large language model (LLM) services become widely adopted, the cost of GPU resources for serving these models in cloud environments has emerged as a critical concern. Spot instances offer up to 90% cost savings over on-demand instances, but their frequent interruptions and limited availability pose significant challenges for continuous LLM serving. GPU spot instances, in particular, exhibit lower and more volatile availability than CPU-based instances, making homogeneous clusters that depend on a single GPU type vulnerable to correlated failures. Heterogeneous clusters spanning multiple GPU types can address this by leveraging complementary availability patterns across diverse spot pools, yet existing LLM serving systems are designed for homogeneous environments and suffer from load imbalance when deployed on heterogeneous GPUs. This paper presents ShuntServe, a cost-efficient LLM serving system for heterogeneous spot GPU clusters. ShuntServe employs a roofline model-based analytical serving performance estimator and a dynamic programming-based model placement optimizer that jointly determines node configuration, parallelization strategy, and layer assignment to maximize throughput across heterogeneous GPUs. To enhance fault tolerance when using spot instances, ShuntServe combines output-preserving request migration with concurrent initialization via a shared tensor store, minimizing migration downtime by overlapping replacement node preparation with ongoing serving. Evaluation on Llama-3.1-70B and Qwen3-32B with a heterogeneous AWS cluster of L4, A10G, and L40S GPUs shows that ShuntServe achieves 1.42x and 1.35x higher throughput than state-of-the-art baselines and attains 31.9% and 31.2% cost efficiency improvements over on-demand instances for offline and online serving, respectively.

cs.DC↗

arXiv2026

SpotVista: Availability-Aware Recommendation System for Reliable and Cost-Efficient Multi-Node Spot Instances

Cloud vendors offer discounted spot instances to maximize surplus resource utilization, but these instances are subject to the risk of sudden interruption. Traditional pricing datasets have been employed to predict this risk, yet recent policy changes by cloud vendors have diminished their effectiveness. To promote spot instance usage, public cloud vendors provide instant availability datasets to help users mitigate interruption risks. While existing research utilizing this data has proposed methods to reduce interruptions, these studies have primarily focused on single-node instances, overlooking the stability of multi-node environments widely adopted for modern cloud workloads. This paper proposes SpotVista, a system that recommends a resource pool of reliable and cost-efficient multi-node spot instances by leveraging various publicly available datasets. To achieve this, SpotVista collects a large-scale multi-node availability dataset while overcoming significant query limitations. Through a thorough analysis of multi-node spot instance availability behavior, SpotVista establishes a methodology for recommending cost-efficient and reliable multi-node configurations. To evaluate how effectively the proposed methodology reflects multi-node availability and cost efficiency, extensive real-world interruption experiments were conducted. The results demonstrate that SpotVista outperforms the state-of-the-art work, SpotVerse, achieving 81.28% greater availability and 2.84\% more cost savings in a multi-region setup. When compared to a publicly available service, AWS SpotFleet, SpotVista provides 21.6\% higher stability and 26.3% greater cost savings.

cs.DC↗

arXiv2026

Permutation-Equivariant 2D State Space Models: Theory and Canonical Architecture for Multivariate Time Series

Multivariate time series (MTS) modeling often implicitly imposes an artificial ordering over variables, violating the inherent exchangeability found in many real-world systems where no canonical variable axis exists. We formalize this limitation as a violation of the permutation symmetry principle and require state-space dynamics to be permutation-equivariant along the variable axis. In this work, we theoretically characterize the complete canonical form of linear variable coupling under this symmetry constraint. We prove that any permutation-equivariant linear 2D state-space system naturally decomposes into local self-dynamics and a global pooled interaction, rendering ordered recurrence not only unnecessary but structurally suboptimal. Motivated by this theoretical foundation, we introduce the Variable-Invariant Two-Dimensional State Space Model (VI 2D SSM), which realizes the canonical equivariant form via permutation-invariant aggregation. This formulation eliminates sequential dependency chains along the variable axis, reducing the dependency depth from $\mathcal{O}(C)$ to $\mathcal{O}(1)$ and simplifying stability analysis to two scalar modes. Furthermore, we propose VI 2D Mamba, a unified architecture integrating multi-scale temporal dynamics and spectral representations. Extensive experiments on forecasting, classification, and anomaly detection benchmarks demonstrate that our model achieves state-of-the-art performance with superior structural scalability, validating the theoretical necessity of symmetry-preserving 2D modeling.

stat.ML↗

arXiv2025

RoboLoc: A Benchmark Dataset for Point Place Recognition and Localization in Indoor-Outdoor Integrated Environments

Robust place recognition is essential for reliable localization in robotics, particularly in complex environments with frequent indoor-outdoor transitions. However, existing LiDAR-based datasets often focus on outdoor scenarios and lack seamless domain shifts. In this paper, we propose RoboLoc, a benchmark dataset designed for GPS-free place recognition in indoor-outdoor environments with floor transitions. RoboLoc features real-world robot trajectories, diverse elevation profiles, and transitions between structured indoor and unstructured outdoor domains. We benchmark a variety of state-of-the-art models, point-based, voxel-based, and BEV-based architectures, highlighting their generalizability domain shifts. RoboLoc provides a realistic testbed for developing multi-domain localization systems in robotics and autonomous navigation

cs.RO↗

arXiv2023

Deep Efficient Continuous Manifold Learning for Time Series Modeling

Modeling non-Euclidean data is drawing extensive attention along with the unprecedented successes of deep neural networks in diverse fields. Particularly, a symmetric positive definite matrix is being actively studied in computer vision, signal processing, and medical image analysis, due to its ability to learn beneficial statistical representations. However, owing to its rigid constraints, it remains challenging to optimization problems and inefficient computational costs, especially, when incorporating it with a deep learning framework. In this paper, we propose a framework to exploit a diffeomorphism mapping between Riemannian manifolds and a Cholesky space, by which it becomes feasible not only to efficiently solve optimization problems but also to greatly reduce computation costs. Further, for dynamic modeling of time-series data, we devise a continuous manifold learning method by systematically integrating a manifold ordinary differential equation and a gated recurrent neural network. It is worth noting that due to the nice parameterization of matrices in a Cholesky space, training our proposed network equipped with Riemannian geometric metrics is straightforward. We demonstrate through experiments over regular and irregular time-series datasets that our proposed model can be efficiently and reliably trained and outperforms existing manifold methods and state-of-the-art methods in various time-series tasks.

cs.LG↗

arXiv2023

Deep Geometric Learning with Monotonicity Constraints for Alzheimer's Disease Progression

Alzheimer's disease (AD) is a devastating neurodegenerative condition that precedes progressive and irreversible dementia; thus, predicting its progression over time is vital for clinical diagnosis and treatment. Numerous studies have implemented structural magnetic resonance imaging (MRI) to model AD progression, focusing on three integral aspects: (i) temporal variability, (ii) incomplete observations, and (iii) temporal geometric characteristics. However, deep learning-based approaches regarding data variability and sparsity have yet to consider inherent geometrical properties sufficiently. The ordinary differential equation-based geometric modeling method (ODE-RGRU) has recently emerged as a promising strategy for modeling time-series data by intertwining a recurrent neural network and an ODE in Riemannian space. Despite its achievements, ODE-RGRU encounters limitations when extrapolating positive definite symmetric metrics from incomplete samples, leading to feature reverse occurrences that are particularly problematic, especially within the clinical facet. Therefore, this study proposes a novel geometric learning approach that models longitudinal MRI biomarkers and cognitive scores by combining three modules: topological space shift, ODE-RGRU, and trajectory estimation. We have also developed a training algorithm that integrates manifold mapping with monotonicity constraints to reflect measurement transition irreversibility. We verify our proposed method's efficacy by predicting clinical labels and cognitive scores over time in regular and irregular settings. Furthermore, we thoroughly analyze our proposed framework through an ablation study.

cs.AI↗

arXiv2022

Layered Cost-Map-Based Traffic Management for Multiple Automated Mobile Robots via a Data Distribution Service

This letter proposes traffic management for multiple automated mobile robots (AMRs) based on a layered cost map. Multiple AMRs communicate via a data distribution service (DDS), which is shared by topics in the same DDS domain. The cost of each layer is manipulated by topics. The traffic management server in the domain sends or receives topics to each of AMRs. Using the layered cost map, the new concept of prohibition filter, lane filter, fleet layer, and region filter are proposed and implemented. The prohibition filter can help a user set an area that would prohibit an AMR from trespassing. The lane filter can help set one-way directions based on an angle image. The fleet layer can help AMRs share their locations via the traffic management server. The region filter requests for or receives an exclusive area, which can be occupied by only one AMR, from the traffic management server. All the layers are experimentally validated with real-world AMRs. Each area can be configured with user-defined images or text-based parameter files.

cs.RO↗

arXiv2022

Behavior Tree-Based Task Planning for Multiple Mobile Robots using a Data Distribution Service

In this study, we propose task planning framework for multiple robots that builds on a behavior tree (BT). BTs communicate with a data distribution service (DDS) to send and receive data. Since the standard BT derived from one root node with a single tick is unsuitable for multiple robots, a novel type of BT action and improved nodes are proposed to control multiple robots through a DDS asynchronously. To plan tasks for robots efficiently, a single task planning unit is implemented with the proposed task types. The task planning unit assigns tasks to each robot simultaneously through a single coalesced BT. If any robot falls into a fault while performing its assigned task, another BT embedded in the robot is executed; the robot enters the recovery mode in order to overcome the fault. To perform this function, the action in the BT corresponding to the task is defined as a variable, which is shared with the DDS so that any action can be exchanged between the task planning unit and robots. To show the feasibility of our framework in a real-world application, three mobile robots were experimentally coordinated for them to travel alternately to four goal positions by the proposed single task planning unit via a DDS.

cs.RO↗

arXiv2021

Medical Transformer: Universal Brain Encoder for 3D MRI Analysis

Transfer learning has gained attention in medical image analysis due to limited annotated 3D medical datasets for training data-driven deep learning models in the real world. Existing 3D-based methods have transferred the pre-trained models to downstream tasks, which achieved promising results with only a small number of training samples. However, they demand a massive amount of parameters to train the model for 3D medical imaging. In this work, we propose a novel transfer learning framework, called Medical Transformer, that effectively models 3D volumetric images in the form of a sequence of 2D image slices. To make a high-level representation in 3D-form empowering spatial relations better, we take a multi-view approach that leverages plenty of information from the three planes of 3D volume, while providing parameter-efficient training. For building a source model generally applicable to various tasks, we pre-train the model in a self-supervised learning manner for masked encoding vector prediction as a proxy task, using a large-scale normal, healthy brain magnetic resonance imaging (MRI) dataset. Our pre-trained model is evaluated on three downstream tasks: (i) brain disease diagnosis, (ii) brain age prediction, and (iii) brain tumor segmentation, which are actively studied in brain MRI research. The experimental results show that our Medical Transformer outperforms the state-of-the-art transfer learning methods, efficiently reducing the number of parameters up to about 92% for classification and

cs.CV↗

arXiv2020

Multi-Scale Neural network for EEG Representation Learning in BCI

Recent advances in deep learning have had a methodological and practical impact on brain-computer interface research. Among the various deep network architectures, convolutional neural networks have been well suited for spatio-spectral-temporal electroencephalogram signal representation learning. Most of the existing CNN-based methods described in the literature extract features at a sequential level of abstraction with repetitive nonlinear operations and involve densely connected layers for classification. However, studies in neurophysiology have revealed that EEG signals carry information in different ranges of frequency components. To better reflect these multi-frequency properties in EEGs, we propose a novel deep multi-scale neural network that discovers feature representations in multiple frequency/time ranges and extracts relationships among electrodes, i.e., spatial representations, for subject intention/condition identification. Furthermore, by completely representing EEG signals with spatio-spectral-temporal information, the proposed method can be utilized for diverse paradigms in both active and passive BCIs, contrary to existing methods that are primarily focused on single-paradigm BCIs. To demonstrate the validity of our proposed method, we conducted experiments on various paradigms of active/passive BCI datasets. Our experimental results demonstrated that the proposed method achieved performance improvements when judged against comparable state-of-the-art methods. Additionally, we analyzed the proposed method using different techniques, such as PSD curves and relevance score inspection to validate the multi-scale EEG signal information capturing ability, activation pattern maps for investigating the learned spatial filters, and t-SNE plotting for visualizing represented features. Finally, we also demonstrated our method's application to real-world problems.

eess.SP↗