On $π^{*}_{0.6}$: a VLA That Learns From Experience

Here is a latest paper written by several well known ML & AI researchers like Chelsea Finn, Karol Hausmann, Sergey Levine, Jost Tobias Springenberg and others.

Finally, these researchers promise that robots will make an Espresso coffee for you! That's progress we can all enjoy! 😊

From the abstract:

"We study how vision-language-action (VLA) models can improve through real-world deployments via reinforcement learning (RL). We present a general-purpose method, RL with Experience and Corrections via Advantage-conditioned Policies (RECAP), that provides for RL training of VLAs via advantage conditioning.

Our method incorporates heterogeneous data into the self-improvement process, including demonstrations, data from on-policy collection, and expert teleoperated interventions provided during autonomous execution.

RECAP starts by pre-training a generalist VLA with offline RL, which we call , that can then be specialized to attain high performance on downstream tasks through on-robot data collection.

We show that the  model trained with the full RECAP method can fold laundry in real homes, reliably assemble boxes, and make espresso drinks using a professional espresso machine. On some of the hardest tasks, RECAP more than doubles task throughput and roughly halves the task failure rate."

 

[2511.14759] $π^{*}_{0.6}$: a VLA That Learns From Experience