Iris Xu

I'm an undergraduate at MIT studying AI and math. My research focuses on efficient reinforcement learning, scalable ML systems, and intelligent agents that understand and interact with the world.

I am currently a Machine Learning Engineer Intern at ByteDance, working on efficient VLM training, and an undergraduate researcher at MIT CSAIL's Improbable AI Lab. Previously, I've interned at Hudson River Trading and the MIT–IBM Watson AI Lab.

Scholar / GitHub / LinkedIn / Twitter (X)

Research

Learning More from Less: Reinforcement Learning from Hindsight

Iris Xu, Sunshine Jiang, John Marangola, Nitish Dashora, Richard Li, Thomas Liu, Zexue He, Yuheng Zhi, Alex Pentland, Pulkit Agrawal, Zhang-Wei Hong

RSS 2026 Workshop RL4VLA

Project Page / Paper / Code (coming soon)

BOAD method diagram

BOAD: Discovering Hierarchical Software Engineering Agents via Bandit Optimization

Iris Xu, Guangtao Zeng, Zexue He, Charles Jin, Aldo Pareja, Dan Gutfreund, Chuang Gan, Zhang-Wei Hong

ICLR 2026 · ICML 2026 Workshop DL4C

Paper / Code / Talk