FlowDAgger: Robots Learn via Corrections
oier_mees · x · 2026-07-16
Core Conclusion
The authors introduce FlowDAgger: instead of directly fine-tuning a robot foundation model, it leverages human corrective actions to learn how to "steer" an already frozen policy.
Key Methods
- Modern VLA / world-action models struggle to adapt to new robots and tasks.
- Traditional approaches like online imitation learning or retraining often break existing capabilities when data is scarce.
- FlowDAgger uses action inversion to map human corrective actions back into the latent noise space of the frozen generative policy.
- A lightweight controller is then trained using these latent targets for deployment-time adaptation, leaving the base model untouched.
Results
- Requires only 5–20 human interventions to learn.
- Outperforms supervised fine-tuning and latent-space RL in both simulation and real-world robots.
- Applicable to VLA, diffusion policies, world-action models.
- Delivers stable improvements without modifying the pre-trained policy.
Additional Info
This project is a collaboration between Microsoft Research and the University of Washington, with paper, project page, and code available.
Related event: FlowDAgger: Adapting Robot Models via Human Corrections(2 posts)→
More from Embodied
- Humanoid robots are edging closer to the uncanny valley, with warmth and face motion — GlenBradley · 2026-07-21
- Humanoid robots are moving from labs into public culture — Olivier__OG · 2026-07-21
- Polymarket puts Tesla’s California robotaxi launch odds at 16% this year — Polymarket · 2026-07-21
- Tesla expands robotaxi service to Orlando and Tampa — Polymarket · 2026-07-21
- Humanoid robots are approaching a deeper uncanny valley — GlenBradley · 2026-07-21
- Polymarket gives Tesla’s Optimus just a 17% chance of debuting this year — Polymarket · 2026-07-21