Jeremy Siburian

Email: jeremy.siburian [at] weblab.t.u-tokyo.ac.jp

I am a master's student at The University of Tokyo, advised by Yusuke Iwasawa and Yutaka Matsuo. I am currently a visiting researcher at Cornell University, working with Kuan Fang. I have also been fortunate to receive mentorship from Jiayuan Mao.

I have spent time at OMRON SINIC X as a robotics research intern and at Daimler Trucks Asia as a research intern. I received my B.Eng. in Mechanical Engineering from Waseda University, where I was advised by Shigeki Sugano.

News

Sep 2026: 2 papers were accepted to CoRL 2026, see you in Austin!
Jun 2026: 2 papers (PHASE and ViLaIn-TAMP) were accepted to IROS 2026, see you in Pittsburgh!
Apr 2026: I received the Friends of UTokyo (FUTI) scholarship for my visit to Cornell.
Sep 2024: Invited Talk at Tokyo AI (TAI) on "Task and Motion Planning for Long-Horizon Manipulation". See the speaker deck here.
May 2024: Our paper received the Best Video Award at the ICRA 2024 Cooking Robotics Workshop.

Research

I develop learning and planning methods for sample-efficient, generalizable robot manipulation. My central interest is enabling robots to continually learn and self-improve by reusing prior knowledge, adapting efficiently to new tasks, and learning from limited interaction rather than repeatedly training from scratch. More broadly, I am interested in how robots can build useful representations of the physical world and use search, planning, and feedback to decide what to learn and how to improve. I hope to develop physical agents whose capabilities grow over time as they encounter new tasks and environments, with a particular interest in world models, inference-time adaptation, and self-directed learning.