Heesoo Jung 정희수

Ph.D. Student · Sungkyunkwan University
Advisor: Prof. Hogun Park

I am a Ph.D. student specializing in Graph Neural Networks, recommendation systems, and interconnected data, passionate about applying my research to real-world challenges. As an internationally active researcher, I am open to internship opportunities worldwide.

Research interests: Graph Neural Networks, Self-Supervised Learning on Graphs, Knowledge Graph Reasoning, GNN Explainability, Recommendation Systems.

Email Scholar GitHub LinkedIn Twitter

News

June 2026 Received Innovation Felloship scholar from SKKU!
June 2026 Serving as a reviewer for NeurIPS 2026
Feb 2026 Co-authored paper accepted to KDD 2026: "Out-of-Distribution Robust Explainer for Graph Neural Networks". 🎉
Jan 2026 Paper accepted to Expert Systems with Applications (Elsevier) on spatial dependency for domain generalization in time-series data.
Nov 2025 Serving as a reviewer for ICLR 2026, The WebConf 2026, and KDD 2026.
May 2025 Paper accepted to KDD 2025: "Harnessing Influence Function in Explaining Graph Neural Networks". Served as Session Chair!
May 2025 Two papers accepted to ACL 2025 Findings and PAKDD 2025 on knowledge graph reasoning.
Jan 2025 Received Excellent Reviewer Award at KDD 2025.
Jan 2025 Paper accepted to WWW 2025 on balancing graph embedding smoothness in SSL.
Oct 2024 Paper accepted to WSDM 2025: "CIMAGE: Exploiting the Conditional Independence in Masked Graph Auto-encoders".
Sep 2023 Started Ph.D. in Software at Sungkyunkwan University under Prof. Hogun Park.

Selected Publications

Bold = me · ^* = equal contribution · See the full list →

Harnessing Influence Function in Explaining Graph Neural Networks

Heesoo Jung^*, Chanyong Kim^*, Geonhee Han, Hogun Park

Proceedings of the Knowledge Discovery and Data Mining (KDD), 2025

code paper

Balancing Graph Embedding Smoothness in Self-Supervised Learning via Information-Theoretic Decomposition

Heesoo Jung, Hogun Park

Proceedings of the Web Conference (WWW), 2025

code paper

CIMAGE: Exploiting the Conditional Independence in Masked Graph Auto-encoders

Jongwon Park^*, Heesoo Jung^*, Hogun Park

Proceedings of the ACM International Conference on Web Search and Data Mining (WSDM), 2025

code paper

Dual Policy Learning for Aggregation Optimization in Graph Neural Network-based Recommender Systems

Heesoo Jung, Sangpil Kim, Hogun Park

Proceedings of the Web Conference (WWW), 2023

code paper

Education

Sungkyunkwan University (SKKU)

Ph.D. in Software · GPA: 4.19 / 4.5

Advisor: Prof. Hogun Park

Suwon, Republic of Korea
Sep 2023 – present

Services & Reviewing

Session Chair: KDD (2025)
Program Committee: The WebConf (WWW) — short track (2025)
Reviewer: NeurIPS (2025, 2026) · The WebConf (2025, 2026) · KDD (2024, 2025 Excellent Reviewer, 2026) · ICLR (2026)
Scholar & Fellowshipt: Innovation Research Fellowship (SKKU) - (2025, 2026)