frontier models
The Student Learns to Search While Uncertain
On-policy student histories let a privileged teacher train exploration without the teacher needing to explore.
Summary
On-policy student histories let a privileged teacher train exploration without the teacher needing to explore.
ASTEROID lets a partially observed student collect its own history, then asks a privileged expert for actions conditioned on that history. The student therefore learns when to gather information and when to exploit what it found. Across eleven environments, the paper reports roughly double the returns under fixed exploration budgets and about one hundredth the interactions of compared reinforcement-learning baselines. In a vision-denied robot transfer, success rose 40 percent over standard distillation. These are controlled evaluations, not proof of general exploration across open worlds.
Why it matters
On-policy student histories let a privileged teacher train exploration without the teacher needing to explore.
Limits and context
- These are controlled evaluations, not proof of general exploration across open worlds.
Key claims
On-policy student histories let a privileged teacher train exploration without the teacher needing to explore.
Qualification: These are controlled evaluations, not proof of general exploration across open worlds.
Evidence: source-2026-10-22-007
Sources
- arXiv preprint 2610.30105, original v1arXiv · primary research
Corrections
No corrections have been recorded for this story.