Author name cluster

Junho Lee

Possible papers associated with this exact author name in Arrow. This page groups case-insensitive exact name matches and is not a full identity disambiguation profile.

12 papers

2 author rows

TMLR Journal 2026 Journal Article

Is There a Better Source Distribution than Gaussian? Exploring Source Distributions for Image Flow Matching

Junho Lee
Kwanseok Kim
Joonseok Lee

Flow matching has emerged as a powerful generative modeling approach with flexible choices of source distribution. While Gaussian distributions are commonly used, the potential for better alternatives in high-dimensional data generation remains largely unexplored. In this paper, we propose a novel 2D simulation that captures high-dimensional geometric properties in an interpretable 2D setting, enabling us to analyze the learning dynamics of flow matching during training. Based on this analysis, we derive several key insights about flow matching behavior: (1) density approximation can paradoxically degrade performance due to mode discrepancy, (2) directional alignment suffers from path entanglement when overly concentrated, (3) Gaussian's omnidirectional coverage ensures robust learning, and (4) norm misalignment incurs substantial learning costs. Building on these insights, we propose a practical framework that combines norm-aligned training with directionally-pruned sampling. This approach maintains the robust omnidirectional supervision essential for stable flow learning, while eliminating initializations in data-sparse regions during inference. Importantly, our pruning strategy can be applied to any flow matching model trained with a Gaussian source, providing immediate performance gains without the need for retraining. Empirical evaluations demonstrate consistent improvements in both generation quality and sampling efficiency. Our findings provide practical insights and guidelines for source distribution design and introduce a readily applicable technique for improving existing flow matching models. Our code is available at https://github.com/kwanseokk/SourceFM.

PDF Details

IJCAI Conference 2025 Conference Paper

Classifying and Tracking International Aid Contribution Towards SDGs

Sungwon Park
Dongjoon Lee
Kyeongjin Ahn
Yubin Choi
Junho Lee
Meeyoung Cha
Kyung Ryul Park

International aid is a critical mechanism for promoting economic growth and well-being in developing nations, supporting progress toward the Sustainable Development Goals (SDGs). However, tracking aid contributions remains challenging due to labor-intensive data management, incomplete records, and the heterogeneous nature of aid data. Recognizing the urgency of this challenge, we partnered with government agencies to develop an AI model that complements manual classification and mitigates human bias in subjective interpretation. By integrating SDG-specific semantics and leveraging prior knowledge from language models, our approach enhances classification accuracy and accommodates the diversity of aid projects. When applied to a comprehensive dataset spanning multiple years, our model can reveal hidden trends in the temporal evolution of international development cooperation. Expert interviews further suggest how these insights can empower policymakers with data-driven decision-making tools, ultimately improving aid effectiveness and supporting progress toward SDGs.

PDF Details DOI

EAAI Journal 2025 Journal Article

Korean football in-game conversation state tracking dataset for dialogue and turn level evaluation

Sangmin Song
Juhyoung Park
Juhwan Choi
Junho Lee
Kyohoon Jin
YoungBin Kim

Recent research in dialogue state tracking has made significant progress in tracking user goals through dialogue-level and turn-level approaches, but existing research primarily focused on predicting dialogue-level belief states. In this study, we present the KICK: Korean football In-game Conversation state tracKing dataset, which introduces a conversation-based approach. This approach leverages the roles of casters and commentators within the self-contained context of sports broadcasting to examine how utterances impact the belief state at both the dialogue-level and turn-level. Towards this end, we propose a task that aims to track the states of a specific time turn and understand conversations during the entire game. The proposed dataset comprises 228 games and 2463 events over one season, with a larger number of tokens per dialogue and turn, making it more challenging than existing datasets. Experiments revealed that the roles and interactions of casters and commentators are important for improving the zero-shot state tracking performance. By better understanding role-based utterances, we identify distinct approaches to the overall game process and events at specific turns.

Details DOI

ICML Conference 2024 Conference Paper

Isometric Representation Learning for Disentangled Latent Space of Diffusion Models

Jaehoon Hahm
Junho Lee
Sunghyun Kim
Joonseok Lee

The latent space of diffusion model mostly still remains unexplored, despite its great success and potential in the field of generative modeling. In fact, the latent space of existing diffusion models are entangled, with a distorted mapping from its latent space to image space. To tackle this problem, we present Isometric Diffusion, equipping a diffusion model with a geometric regularizer to guide the model to learn a geometrically sound latent space of the training data manifold. This approach allows diffusion models to learn a more disentangled latent space, which enables smoother interpolation, more accurate inversion, and more precise control over attributes directly in the latent space. Our extensive experiments consisting of image interpolations, image inversions, and linear editing show the effectiveness of our method.

Details

NeurIPS Conference 2024 Conference Paper

Self-Guided Masked Autoencoder

Jeongwoo Shin
Inseo Lee
Junho Lee
Joonseok Lee

Masked Autoencoder (MAE) is a self-supervised approach for representation learning, widely applicable to a variety of downstream tasks in computer vision. In spite of its success, it is still not fully uncovered what and how MAE exactly learns. In this paper, with an in-depth analysis, we discover that MAE intrinsically learns pattern-based patch-level clustering from surprisingly early stages of pre-training. Upon this understanding, we propose self-guided masked autoencoder, which internally generates informed mask by utilizing its progress in patch clustering, substituting the naive random masking of the vanilla MAE. Our approach significantly boosts its learning process without relying on any external models or supplementary information, keeping the benefit of self-supervised nature of MAE intact. Comprehensive experiments on various downstream tasks verify the effectiveness of the proposed method.

PDF Details DOI

IJCAI Conference 2023 Conference Paper

Machine Learning Driven Aid Classification for Sustainable Development

Junho Lee
Hyeonho Song
Dongjoon Lee
Sundong Kim
Jisoo Sim
Meeyoung Cha
Kyung-Ryul Park

This paper explores how machine learning can help classify aid activities by sector using the OECD Creditor Reporting System (CRS). The CRS is a key source of data for monitoring and evaluating aid flows in line with the United Nations Sustainable Development Goals (SDGs), especially SDG17 which calls for global partnership and data sharing. To address the challenges of current labor-intensive practices of assigning the code and the related human inefficiencies, we propose a machine learning solution that uses ELECTRA to suggest relevant five-digit purpose codes in CRS for aid activities, achieving an accuracy of 0. 9575 for the top-3 recommendations. We also conduct qualitative research based on semi-structured interviews and focus group discussions with SDG experts who assess the model results and provide feedback. We discuss the policy, practical, and methodological implications of our work and highlight the potential of AI applications to improve routine tasks in the public sector and foster partnerships for achieving the SDGs.

PDF Details DOI

NeurIPS Conference 2023 Conference Paper

SiT Dataset: Socially Interactive Pedestrian Trajectory Dataset for Social Navigation Robots

Jong Wook Bae
Jungho Kim
Junyong Yun
Changwon Kang
Jeongseon Choi
Chanhyeok Kim
Junho Lee
Jungwook Choi

To ensure secure and dependable mobility in environments shared by humans and robots, social navigation robots should possess the capability to accurately perceive and predict the trajectories of nearby pedestrians. In this paper, we present a novel dataset of pedestrian trajectories, referred to as Social Interactive Trajectory (SiT) dataset, which can be used to train pedestrian detection, tracking, and trajectory prediction models needed to design social navigation robots. Our dataset includes sequential raw data captured by two 3D LiDARs and five cameras covering a 360-degree view, two inertial measurement unit (IMU) sensors, and real-time kinematic positioning (RTK), as well as annotations including 2D & 3D boxes, object classes, and object IDs. Thus far, various human trajectory datasets have been introduced to support the development of pedestrian motion forecasting models. Our SiT dataset differs from these datasets in the following two respects. First, whereas the pedestrian trajectory data in other datasets was obtained from static scenes, our data was collected while the robot navigates in a crowded environment, capturing human-robot interactive scenarios in motion. Second, our dataset has been carefully organized to facilitate training and evaluation of end-to-end prediction models encompassing 3D detection, 3D multi-object tracking, and trajectory prediction. This design allows for an end-to-end unified modular approach across different tasks. We have introduced a comprehensive benchmark for assessing models across all aforementioned tasks, and have showcased the performance of multiple baseline models as part of our evaluation. Our dataset provides a strong foundation for future research in pedestrian trajectory prediction, which could expedite the development of safe and agile social navigation robots. The SiT dataset, devkit, and pre-trained models are publicly released at: https: //spalaboratory. github. io/SiT

PDF Details

ICLR Conference 2022 Conference Paper

Do Not Escape From the Manifold: Discovering the Local Coordinates on the Latent Space of GANs

Jaewoong Choi
Junho Lee
Changyeon Yoon
Jung Ho Park
Geonho Hwang
Myungjoo Kang

The discovery of the disentanglement properties of the latent space in GANs motivated a lot of research to find the semantically meaningful directions on it. In this paper, we suggest that the disentanglement property is closely related to the geometry of the latent space. In this regard, we propose an unsupervised method for finding the semantic-factorizing directions on the intermediate latent space of GANs based on the local geometry. Intuitively, our proposed method, called $\textit{Local Basis}$, finds the principal variation of the latent space in the neighborhood of the base latent variable. Experimental results show that the local principal variation corresponds to the semantic factorization and traversing along it provides strong robustness to image traversal. Moreover, we suggest an explanation for the limited success in finding the global traversal directions in the latent space, especially $\mathcal{W}$-space of StyleGAN2. We show that $\mathcal{W}$-space is warped globally by comparing the local geometry, discovered from Local Basis, through the metric on Grassmannian Manifold. The global warpage implies that the latent space is not well-aligned globally and therefore the global traversal directions are bound to show limited success on it.

Details

IROS Conference 2022 Conference Paper

MasKGrasp: Mask-based Grasping for Scenes with Multiple General Real-world Objects

Junho Lee
Junhwa Hur
Inwoo Hwang
Young Min Kim 0001

In this paper, we introduce a mask-based grasping method that discerns multiple objects within the scene regard-less of transparency or specularity and finds the optimal grasp position avoiding clutter. Conventional vision-based robotic grasping approaches often fail to extend to the scenes containing transparent objects due to their different visual appearance. To handle the different visual characteristics, we first segment both transparent and opaque objects into instance masks, which serve as the domain-agnostic intermediate representation of both object types, using a neural network. While there exists no labelled training dataset that strongly represents both object types, we overcome the limitation by augmenting transparent objects on an existing large-scale dataset. Then, given the object instance masks, our method selects the top K discrete masks and robustly estimates grasp poses avoiding clutter. Through experiments, we verify that the instance masks are light-weight yet provide sufficient information for vision-based grasping agnostic of various appearances. On an unseen real-world test environment with complex objects, our method substantially outperforms previous methods without fine-tuning.

Details

ICAPS Conference 2019 Conference Paper

Backward Sequence Analysis for Single-Armed Cluster Tools

Junho Lee
Hyun-Jung Kim

This paper analyzes a backward sequence for single-armed cluster tools with processing time variations. A fundamental cycle is defined with the backward sequence, and a formula for the cycle time is derived by considering processing time variations. Then conditions for which the backward sequence is optimal are developed. The upper bound on the average cycle time from the backward sequence is also analyzed. We then show experimentally that the sequence performs well even with processing time variations.

Details

ICRA Conference 2018 Conference Paper

Completion Time Analysis for Automated Manufacturing Systems with Parallel Processing Modules

Hyun-Jung Kim
Junho Lee

This paper analyzes the completion time of automated manufacturing systems, especially a dual-armed cluster tool, equipped with parallel processing modules (PMs). Cluster tools, which consist of multiple PMs, a transport robot, and loadlocks where wafer lots are loaded and unloaded, perform semiconductor manufacturing processes, such as lithography, etching, deposition, and testing. Wafer lots are transported by overhead hoist transports (OHTs) between tools or stockers where wafer lots are stored. To reduce the idle time of OHTs or cluster tools, it is essential to estimate the time when all wafers of a lot finish processing in a tool. Hence, we derive closed-form expressions for the completion time of wafer lots, especially in dual-armed cluster tools with parallel PMs to reflect real circumstances of fabs. We finally show that the formulas derived can be used even when there are small processing time variations with numerical experiments.

Details

ICRA Conference 2012 Conference Paper

Scheduling transient periods of single-armed cluster tools

Junho Lee
Tae-Eog Lee

Semiconductor manufacturing fabs recently tend to reduce the lot size, that is, the number of identical wafers in a lot, because of small lot orders and increased die throughput per wafer due to wafer size increase. Therefore, cluster tools for wafer processing, which mostly repeat identical work cycles, are subject to frequent lot changes. We therefore examine scheduling problems for transient periods of single-armed cluster tools that are scheduled to repeat identical work cycles for a number of identical wafers. We first develop a Petri net model for the tool's operational behavior including the initial transient periods as well as the steady cycles. We then develop a mixed integer programming model for finding an optimal schedule. We also examine how to adapt the simple backward sequence, which is mostly used for scheduling steady work cycles of single-armed cluster tools, for a transient period. We identify a deadlock-free condition and also propose two efficient heuristic algorithms by modifying the backward sequence. Finally, through computational experiments, we analyze the efficiency of the proposed algorithms.

Details