Profile Photo

Dongsu Lee

Hello! I am a PhD student at the University of Texas at Austin, where I’m supervised by Prof. Amy Zhang in the MIDI Lab. Previous affiliation. I collaborated with Prof. Ding Zhao in the Safe AI Lab, Carnegie Mellon University (CMU); with Prof. Minhae Kwon in the BMIL; and Prof. Sangsoo Kim in the Post-genome Informatics Lab at Soongsil University in Seoul, South Korea.

I’m interested in realistic RL problems, e.g, partial observability, coordination with unseen partners, and human-robot collaboration. Then, I aim to make an RL algorithm applicable in via both self-play and data-driven techniques.

I am always happy to talk about research, so feel free to reach out if you want to chat!

Selected Publications

Partner as Task: Belief-conditioned Successor Features for Zero-shot Coordination
Dongsu Lee, Lucas Maes, Quentin Le Lidec, Randall Balestriero, Amy Zhang
NeurIPS 2026 PTA Workshop
Beyond Training Time, Test-Time Coordination is Essential for Cooperative MARL
Dongsu Lee*, Sooraj Sathish*, Joonkyung Kim, Woojun Kim, Amy Zhang
NeurIPS Position Track 2026
Multi-agent Coordination via Flow Matching
Dongsu Lee, Daehee Lee, Amy Zhang
ICLR 2026

All Publications

2026
Partner as Task: Belief-conditioned Successor Features for Zero-shot Coordination
Dongsu Lee, Lucas Maes, Quentin Le Lidec, Randall Balestriero, Amy Zhang
NeurIPS 2026 PTA Workshop
Beyond Training Time, Test-Time Coordination is Essential for Cooperative MARL
Dongsu Lee*, Sooraj Sathish*, Joonkyung Kim, Woojun Kim, Amy Zhang
NeurIPS Position Track 2026
Reconfiguring Procedural Knowledge for Compositional Robot Skill Adaptation
Daehee Lee, Dongsu Lee, Sanghyun Ahn, Minjong Yoo, Honguk Woo
NeurIPS 2026
Multi-agent Coordination via Flow Matching
Dongsu Lee, Daehee Lee, Amy Zhang
ICLR 2026
2025
Unifying Agent Interaction and World Information for Multi-agent Coordination
Dongsu Lee, Daehee Lee, Yaru Niu, Honguk Woo, Amy Zhang, Ding Zhao
NeurIPS 2025 ARLET Workshop (Oral)
Policy Compatible Skill Incremental Learning via Lazy Learning Interface
Daehee Lee, Dongsu Lee, TaeYoon Kwack, Wonje Choi, Honguk Woo
NeurIPS 2025 (Spotlight)
Scenario-free Autonomous Driving with Multi-task Offline-to-online Reinforcement Learning
Dongsu Lee, Minhae Kwon
IEEE Transactions on Intelligent Transportation Systems
2024

Write-ups

Blog A lazy, borrowed recipe
Mar 2026 • Discourse: right direction of the offline MARL community.
Blog Fast and expressive multi-agent coordination
Oct 2025 • Capturing complex joint behavior while keeping inference fast.
Blog Research statement: Towards generalizable MARL
Aug 2025 • Bottlenecks and a roadmap for generalizable MARL.