Home /Research /FlowDAgger: Human-in-the-Loop Adaptation of Generative Robot Policies in Latent Space
LEARNING

FlowDAgger: Human-in-the-Loop Adaptation of Generative Robot Policies in Latent Space

Michael Murray, Daphne Chen, Simran Bagaria, Dean Fortier, Tess Hellebrekers, Galen Mullins, Harshavardhan Gajarla, Oier Mees, Maya Cakmak, Andrey Kolobov

Year
2026
Access
Open access

Abstract

Pretrained generative robot policies based on flow matching and diffusion have achieved impressive results across a wide range of manipulation tasks. Yet real-world deployments routinely expose failure modes outside the pretraining distribution. Closing these gaps typically requires large-scale data collection or online reinforcement learning on physical hardware, which is impractical for rapid and safe adaptation. We present FlowDAgger, a sample- and compute-efficient method for adapting frozen generative robot policies from human interventions in latent space. Our 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. The resulting inverted noise provides supervision for a lightweight latent policy that steers the base model at deployment time, enabling rapid skill acquisition while preserving its behavioral priors. We evaluate FlowDAgger in simulation and on real-world bimanual and single-arm manipulation, adapting both action-head VLAs and world-action models from a handful of interventions. FlowDAgger outperforms supervised fine-tuning and latent-space RL baselines and preserves pretrained skills on held-out tasks, offering a practical path for adapting robot foundation models in the real world. Website: https://microsoft.github.io/FlowDAgger

Keywords

human-in-the-loopgenerative policylatent space adaptationaction inversionflow matching

Related papers

Browse all LEARNING papers