Florent Lamiraux

MAGMA-GEN is an on-policy data-generation pipeline that converts ambiguous failed rollouts into validated recovery supervision and improves task success and recovery capabilities, against distillation and trajectory-repair baselines under evolving task constraints in both simulation and real-robot execution.

Robots can now learn from prior experience to deburr parts more reliably and safely, even in cluttered environments, by combining diffusion models with force-feedback MPC.