Jiho Choi

Hello. 👋
I am a Ph.D. student in the Graduate School of AI at KAIST, fortunate to be advised by Prof. Hyunjung Shim. Previously, I worked as a Data Scientist at Netmarble AI Center (2020-2023), focusing on anomaly detection and fraud detection for online games. I received my M.S. in Computer Science and Engineering from Seoul National University (2018-2020), where I was grateful to be advised by Prof. Chong-kwon Kim. I received my B.S. in Computer Science from Purdue University, West Lafayette (2013-2017). My research interests include multimodal learning, vision-language models, open-vocabulary, and graph neural networks.

Curriculum Vitae (CV)

📚

Education

KAIST, Seoul, Republic of Korea

Ph.D. Student in Artificial Intelligence (Graduate School of AI), Feb. 2024 - Present

Advisor: Prof. Hyunjung Shim
News:

Seoul National University, Seoul, Republic of Korea

M.S. in Computer Science & Engineering, Aug. 2020

Advisor: Chong-kwon Kim
M.S. Thesis: Dynamic Graph Convolutional Networks with Attention Mechanism for Rumor Detection on Social Media

Purdue University, West Lafayette, IN, USA

B.S. in Computer Science, Dec. 2017

Individual Research: Cloud Computing Security

💼

Experience

Data Scientist

Netmarble Corp., Seoul, South Korea, Sep. 2020 - Dec. 2023

Data Science Division / Abnormal User Information Team
News: , | Patents: ,

Intern

NAVER LABS Corp., Seongnam, South Korea, Jul. 2019 - Jan. 2020

Autonomous Driving Group / Mapping & Localization Team
Slides: | Demo:

Intern

Fasoo Inc., Seoul, South Korea, May 2017 - Aug. 2017

Sparrow Static Analyzer Development Team

🏆

Awards & Honors

CVPR 2025 - Outstanding Reviewer (Top 5.6% (≈ 711 / 12,592))
2nd Place in Track 1, Open-Vocabulary Part Segmentation Challenge - VPLOW @ CVPR 2024
2nd Place in Track 2, Open-Vocabulary Part Segmentation Challenge - VPLOW @ CVPR 2024
Teaching / Research Assistant Scholarship - Brain Korea 21 Plus, Seoul National University (2018 - 2020)
Best Presentation Award - Korea Computer Congress 2019
Merit-based Scholarship - Seoul National University (Spring 2018)
SOGA Grant - Purdue Korean Undergraduate Student Association (2015 - 2016)

🔬

Research Interests

Multimodal Learning
Vision-Language Models
Open-Vocabulary Segmentation
Graph Mining & Graph Neural Networks

📝

Publications

(* indicates equal contributions)

Preprints

PosterForest: Hierarchical Multi-Agent Collaboration for Scientific Poster Generation

Jiho Choi*, Seojeong Park*, Seongjong Song, Hyunjung Shim

2025.08

MomentMix Augmentation with Length-Aware DETR for Temporally Robust Moment Retrieval

Seojeong Park, Jiho Choi, Kyungjune Baek, Hyunjung Shim

arXiv 2024.12

Refereed Conference Publications

3D-Aware VLMs with Geometric Distillation

3D-Aware Vision-Language Models Fine-Tuning with Geometric Distillation

Seonho Lee*, Jiho Choi*, Inha Kang, Jiwook Kim, Junsung Park, Hyunjung Shim

EMNLP 2025 Findings (Acceptance Rate: 39.5% (≈ (1,811 + 1,418) / 8,174) (Main + Findings))

Scribble-Guided Diffusion for Training-free Text-to-Image Generation

Seonho Lee*, Jiho Choi*, Seohyun Lim, Jiwook Kim, Hyunjung Shim

ICIP 2025 (Acceptance Rate: 38.2% (≈ 518 / 1,355))

Selective Blocking Framework for Message-Passing Neural Networks on Heterophilic and Uncertain Graphs

Younhyuk Choi, Taewook Ko, Jiho Choi, Chong-kwon Kim

UAI 2025 (Acceptance Rate: 30.7% (≈ 230 / 750))

Fine-Grained Image-Text Correspondence with Cost Aggregation for Open-Vocabulary Part Segmentation

Jiho Choi*, Seonho Lee*, Minhyun Lee, Seungho Lee, Hyunjung Shim

CVPR 2025 (Acceptance Rate: 22.1% (≈ 2,878 / 13,008))

DreamCatalyst: Fast and High-Quality 3D Editing via Controlling Editability and Identity Preservation

Jiwook Kim*, Seonho Lee*, Jaeyo Shin, Jiho Choi, Hyunjung Shim

ICLR 2025 (Acceptance Rate: 32.1% (≈ 3,710 / 11,565))

Mitigating Overfitting in Graph Neural Networks via Feature and Hyperplane Perturbation

Younhyuk Choi, Jiho Choi, Taewook Ko, Chong-kwon Kim

WSDM 2025 (Acceptance Rate: 17.3% (≈ 106 / 614))

Review-Based Hyperbolic Cross-Domain Recommendation

Younhyuk Choi, Jiho Choi, Taewook Ko, Chong-kwon Kim

WSDM 2025 (Acceptance Rate: 17.3% (≈ 106 / 614))

Understanding Multi-Granularity for Open-Vocabulary Part Segmentation

Jiho Choi*, Seonho Lee*, Seungho Lee, Minhyun Lee, Hyunjung Shim

NeurIPS 2024 (Acceptance Rate: 25.8% (≈ 4,040 / 15,671))

PU GNN: Chargeback Fraud Detection in P2E MMORPGs via Graph Attention Networks with Imbalanced PU Labels

Jiho Choi*, Junghoon Park*, Woocheol Kim, Jin-Hyeok Park, Yumin Suh, Minchang Sung

ECML PKDD 2023 (Applied Data Science Track) (Acceptance Rate: 24.3 % (≈ 58 / 239))

Finding Heterophilic Neighbors via Confidence-based Subgraph Matching for Semi-supervised Node Classification

Younhyuk Choi, Jiho Choi, Taewook Ko, Hyungho Byun, Chong-kwon Kim

CIKM 2022 (Acceptance Rate: 27.5 % (≈ 621 / 2,257))

Review-Based Domain Disentanglement without Duplicate Users or Contexts for Cross-Domain Recommendation

Younhyuk Choi, Jiho Choi, Taewook Ko, Hyungho Byun, Chong-kwon Kim

CIKM 2022 (Acceptance Rate: 27.5 % (≈ 621 / 2,257))

Journal Publications

Beyond Binary: Improving Signed Message Passing in Graph Neural Networks for Multi-Class Graphs

Younhyuk Choi, Taewook Ko, Jiho Choi, Chong-kwon Kim

IEEE TPAMI 2025 (SCIE)

Beyond Message-Passing: Generalization of Graph Neural Networks via Feature Perturbation for Semi-supervised Node Classification

Younhyuk Choi, Jiho Choi, Taewook Ko, Chong-kwon Kim

IEEE TNNLS 2024 (SCIE)

Dynamic Graph Convolutional Networks with Attention Mechanism for Rumor Detection on Social Media

Jiho Choi, Taewook Ko, Younhyuk Choi, Hyungho Byun, Chong-kwon Kim

PLOS ONE 2021 (SCIE)