Dongjun Lee Ph.D student, KAIST AI
I am a third-year Ph.D. student at KAIST, advised by Prof. Kimin Lee, and currently a visiting student at New York University collaborating with Prof. Mengye Ren. Previously, I interned at Config, and also interned at Microsoft Research.
My research goal is robot learning for human-aware, interactive behavior. To this end, I am currently focusing on (i) situated human-robot interaction, (ii) social intelligence and safety, and (iii) human-in-the-loop interactive learning (e.g., RLHF).
Google Scholar | GitHub | LinkedIn | X (Twitter)
dgjun32 [AT] kaist.ac.kr / mando0302 [AT] gmail.com
Publications* indicates equal contribution
![[W2] EDITH: Teaching Robots to Follow Nonverbal Human Intent from Egocentric Vision and Gaze teaser](/assets/papers/edith.gif)
[W2] EDITH: Teaching Robots to Follow Nonverbal Human Intent from Egocentric Vision and Gaze
IROS 2026 Workshop on Human-Robot Dialogue
![[C3] Learning to Generate Unit Test via Adversarial Reinforcement Learning teaser](/assets/papers/utrl.gif)
[C3] Learning to Generate Unit Test via Adversarial Reinforcement Learning
ICLR 2026
![[C2] Learning to Contextualize Web Pages for Enhanced Decision Making by LLM Agents teaser](/assets/papers/lcow.gif)
[C2] Learning to Contextualize Web Pages for Enhanced Decision Making by LLM Agents
ICLR 2025
![[C1] Read-only prompt optimization for Vision-Language Few-shot Learning teaser](/assets/papers/rpo.png)
[C1] Read-only prompt optimization for Vision-Language Few-shot Learning
ICCV 2023
2026
[W2] EDITH: Teaching Robots to Follow Nonverbal Human Intent from Egocentric Vision and Gaze
IROS 2026 Workshop on Human-Robot Dialogue
[C4] Learning Multi-View Spatial Reasoning from Cross-View Relations
CVPR 2026
[C3] Learning to Generate Unit Test via Adversarial Reinforcement Learning
ICLR 2026
[W1] Adversarial Iterative Unit Test Generation with Large Language Models
ICLR 2026 Workshop on Agentic AI in the Wild
2025
[C2] Learning to Contextualize Web Pages for Enhanced Decision Making by LLM Agents
ICLR 2025
[J1] Collaborative LLM Inference via Planning for Efficient Reasoning
TMLR
[P1] Automated Skill Discovery for Language Agents through Exploration and Iterative Feedback
Arxiv Preprint
2023
[C1] Read-only prompt optimization for Vision-Language Few-shot Learning
ICCV 2023
Experience
-
NYU CourantVisiting Scholar (Host: Prof. Mengye Ren) -
ConfigTechnical internWorked on human-robot interaction data pipelines and human-interactive policy training. -
Microsoft ResearchResearch intern (Host: Changho Hwang)Worked on training LLMs via RL for code and unit test generation. -
NAVER CorpResearch internWorked on vision-language representation learning.
Education
-
Korea Advanced Institute of Science and Technology (KAIST)Ph.D. in Artificial Intelligence
-
Korea UniversityB.S. in Statistics & Artificial Intelligence
Academic Services
-
Conference Reviewer, ICLR, CVPR, NeurIPS, CoRL, ICRAJournal Reviewer, Pattern Recognition
Honors & Awards
-
Donggyo Talent Award, Surim Foundation
-
Ilju Academy and Culture Foundation Scholarship, Ilju Foundation
-
Semester High Honor, Korea University