Hoseong Jung

I'm a PhD student in the Interdisciplinary Program in Artificial Intelligence at Seoul National University, working at Lab for Autonomous Robotics Research (LARR) under the guidance of Prof. H. Jin Kim.

My research focuses on cooperative intelligence within multi-agent embodied systems.
Before joining LARR, I was a research officer at the Defense AI Center of the Agency for Defense Development (ADD).
I completed my bachelor's degree in Electrical and Computer Engineering at the Seoul National University.
I hope my research leads to assistant robots that bring care and support to children and people with disabilities.

Email / CV / Bio / Scholar / Twitter / LinkedIn / Github

News

[Sep 2026] Two papers (EFXPLORER, DynaFLIP) are accepted to NeurIPS 2026 .
[Sep 2026] Presented our work (LAMP) at the ITSC 2026 , held in Napoli.
[Jun 2026] Presented our work (TART, LAMP) at the ICRA 2026 , held in Vienna.
[Feb 2026] One paper (TART) is accepted to ICRA 2026 .
[Jun 2025] Presented our work (TART) at the RSS 2025 workshop on Leveraging Implicit Methods for Aerial Autonomy, in LA.
[Oct 2024] Presented our RA-L paper (M-DT) and late breaking results at IROS 2024, held in Abu Dhabi.
[Mar 2024] One paper is accepted to IEEE Robotics and Automation Letters (RA-L).

Selected Publications

	Cooperative Multi-Agent Reinforcement Learning via Epigraph-Form Guided Exploration Sungil Son^, Hoseong Jung^, Dahyun Oh^, H. Jin Kim NeurIPS, 2026 project page In this paper, we propose EFXPLORER*, a constrained exploration framework that maximizes exploration objectives subject to a constraint that preserves established task performance.
	LAMP: Lane-Aligned Motion Primitives for Feasible Trajectory Prediction Sangjin Han^, Hoseong Jung^, Jeongtae Her^, Changhyun Choi, H. Jin Kim ICRA workshop on Generalization in Autonomous Driving, 2026 IEEE International Conference on Intelligent Transportation Systems (ITSC)*, 2026 paper / project page In this paper, we propose a topology-aware forecasting framework designed to improve the structural reliability of multimodal predictions across varying road geometries.
	Temporal Action Representation Learning for Tactical Resource Control and Subsequent Maneuver Generation Hoseong Jung, Sungil Son, Daesol Cho, Jonghae Park, Changhyun Choi, H. Jin Kim IEEE International Conference on Robotics and Automation (ICRA), 2026 RSS workshop on Leveraging Implicit Methods for Aerial Autonomy, 2025 paper / project page In this paper, we introduce a temporal action contrastive learning approach that facilitates semantic alignment between resource control and tactical maneuvers.
	Maneuver-Conditioned Decision Transformer for Tactical in-Flight Decision-Making Hoseong Jung, Yong-Duk Kim, Youngjung Kim IEEE Robotics and Automation Letters, 2024 paper In this paper, we explore the offline reinforcement learning framework for tactical air combat.
	Deep Reinforcement Learning-Based Air-to-Air Combat Maneuver Generation in a Realistic Environment Jung Ho Bae, Hoseong Jung, Seogbong Kim, Sungho Kim, Yong-Duk Kim IEEE Access, 2023 paper In this paper, we propose a deep reinforcement learning-based framework for developing a model capable of performing within visual range (WVR) air-to-air combat under the conditions of a partially observable Markov decision process (POMDP) with insufficient information.

Projects

AI for Air Combat Engagement
Agency for Defense Development, 2022 - 2023

In this project, we develop both heuristic-based and learning-based agent performing within visual range (WVR) air-to-air combat.

Awards

[Scholarship] Hyundai Motor Chung Mong-Koo Foundation (Aug 2025 – Present)

Feel free to steal this website's source code. Do not scrape the HTML from this page itself, as it includes analytics tags that you do not want on your own website — use the github code instead. Also, consider using Leonid Keselman's Jekyll fork of this page.