Preprint
Jul 2026
FlowDAgger: Human-in-the-Loop Adaptation of Generative Robot Policies in Latent Space
The key idea is action inversion: each human expert action is mapped to the noise that would have produced it under the frozen base policy, using reverse-time integration followed by local refinement, which provides supervision for a lightweight latent policy that steers the base model at deployment time, enabling rapid skill acquisition while preserving its behavioral priors.
Michael Murray, Daphne Chen, Simran Bagaria et al.
· 1 citation