Maximilian Xiling Li

Research

My research focuses on enabling robots to learn and understand manipulation skills. I work on imitation learning — developing methods that allow robots to efficiently acquire complex, versatile behaviors from demonstrations. A central theme in my work is interpretable AI: I investigate prototype-based and other transparent model architectures that make the decision processes of learned policies legible to humans. On the perception side, I work on 3D robot perception, including affordance detection on point clouds, to ground these learned skills in rich geometric representations of the environment.

Background

Max studied Computer Science at TU Darmstadt, during which he also worked as a research working student at SAP. He continued at KIT for his master's, first as a research assistant at the Human-Centered Systems Lab, then at the Autonomous Learning Robots Lab. He is now a PhD researcher at the Intuitive Robots Lab.

Selected Publications

• Multi-Objective Photoreal Simulation (MOPS) Dataset for Computer Vision in Robotic Manipulation Maximilian Xiling Li, Paul Mattes, Nils Blank, Korbinian Franz Rudolf, Paul Werner Lödige, Rudolf Lioutikov Workshop on Structured World Models for Robotic Manipulation, 2025
  Project
• Use the Force, Bot! — Force-aware ProDMP with Event-based Replanning Paul Werner Lödige, Maximilian Xiling Li, Rudolf Lioutikov ICRAIEEE International Conference on Robotics and Automation, 2025
  Project arXiv Code
• Information Maximizing Curriculum: A Curriculum-based Approach for Learning Versatile Skills Denis Blessing, Onur Celik, Xiaogang Jia, Moritz Reuss, Maximilian Li, Rudolf Lioutikov, Gerhard Neumann NeurIPSAdvances in Neural Information Processing Systems, 2023
  Paper arXiv
• Goal Conditioned Imitation Learning using Score-based Diffusion Policies Moritz Reuss, Maximilian Li, Xiaogang Jia, Rudolf Lioutikov RSSRobotics: Science and Systems, 2023
  Project arXiv Code

View all publications →