Make π0.5 Learn to Call Astra for Reliable Robot Manipulation
Overview
A system combines π0.5 with Astra to improve robot manipulation success rates. Astra is used selectively during critical events, achieving 97.5% success in 40 LIBERO tasks, surpassing π0.5's 95.5% baseline. However, frequent Astra queries increase inference time and costs.
Generated from attributed reports · Updated 54 minutes ago
Event evidence and corrections
0 attributed source owners. Ownership does not establish independent confirmation. Quantities are reported separately and are never added together.
Reported quantity · Reported success rate: 97.5 percent · Basis not reported
“solely for critical recovery events and returns control to $\pi_{0.5}$ upon stabilization. Evaluated across 40 LIBERO tasks (200 episodes), our system achieves a 97.5% success rate, outperforming the 95.5% baseline of pure $\pi_{0.5}$. In the Object, Goal, and Spatial suites, Astra executes only 5.49% of environment steps, consuming approximately 5,027 tokens per test episode. This selective intervention preserves t”
Exact source · revision 1Source owner not reported
Reported quantity · tokens per test episode: 5027 other · Basis not reported
“solely for critical recovery events and returns control to $\pi_{0.5}$ upon stabilization. Evaluated across 40 LIBERO tasks (200 episodes), our system achieves a 97.5% success rate, outperforming the 95.5% baseline of pure $\pi_{0.5}$. In the Object, Goal, and Spatial suites, Astra executes only 5.49% of environment steps, consuming approximately 5,027 tokens per test episode. This selective intervention preserves t”
Exact source · revision 1Source owner not reported
Reported quantity · episodes: 200 other · Basis not reported
“solely for critical recovery events and returns control to $\pi_{0.5}$ upon stabilization. Evaluated across 40 LIBERO tasks (200 episodes), our system achieves a 97.5% success rate, outperforming the 95.5% baseline of pure $\pi_{0.5}$. In the Object, Goal, and Spatial suites, Astra executes only 5.49% of environment steps, consuming approximately 5,027 tokens per test episode. This selective intervention preserves t”
Exact source · revision 1Source owner not reported
Reported quantity · tasks: 40 other · Basis not reported
“solely for critical recovery events and returns control to $\pi_{0.5}$ upon stabilization. Evaluated across 40 LIBERO tasks (200 episodes), our system achieves a 97.5% success rate, outperforming the 95.5% baseline of pure $\pi_{0.5}$. In the Object, Goal, and Spatial suites, Astra executes only 5.49% of environment steps, consuming approximately 5,027 tokens per test episode. This selective intervention preserves t”
Exact source · revision 1Source owner not reported
Report timeline
Follow attributed reports and material updates.
- arXiv Robotics — research abstractsMake π0.5 Learn to Call Astra for Reliable Robot Manipulation
The paper introduces a system where a frozen π0.5 policy is augmented with a lightweight CALL head that learns to predict when external assistance from Astra is needed. Astra is only activated during critical recovery events, achieving a 97.5% success rate across 40 LIBERO tasks, outperforming the 95.5% baseline of pure π0.5. The approach preserves fast execution while integrating Astra's reasoning on demand.
Timezone · UTC
Article dates follow your selected timezone. Briefing editions use Hong Kong time (UTC+8).