Jean-Francois Lafleche

Papers

Total Citations

H-Index

About

Jean-Francois Lafleche is a leading roboticist whose research bridges the gap between simulation and real-world dexterity, focusing on robot learning, perception, and human-robot collaboration. His most influential work, “Robot Cooperative Behavior Learning Using Single-Shot Learning From Demonstration and Parallel Hidden Markov Models” (2018, 19 citations), introduced a novel PaHMM architecture that enables robots to learn complex collaborative tasks from non-expert users in a single demonstration—a breakthrough for intuitive human-robot interaction. Lafleche further advanced the field with “DeXtreme: Transfer of Agile In-hand Manipulation from Simulation to Reality” (2022, 5 citations), which demonstrated how deep reinforcement learning policies for multi-fingered manipulation can be successfully transferred from simulation to physical robots, overcoming the notorious sim-to-real gap. His recent work, “Synthetica: Large Scale Synthetic Data Generation for Robot Perception” (2025, 5 citations), tackles the critical challenge of generating massive, photorealistic training data for vision-based object detectors, ensuring robust performance under varying lighting and occlusion. Through these contributions, Lafleche has established himself as a key figure in making robots more adaptable, capable, and deployable in real-world environments.