Prompt Breadth and Rollout Refresh Interact in On-Policy Distillation

T1Covered by 1 source · first reported by arXiv at 24 Sept, 04:00 UTC

arXiv:2609.25048v1 Announce Type: cross Abstract: How many prompts does on-policy distillation (OPD) need, and how does the answer depend on the student policies that generate its training responses? We study these two controls jointly: prompt breadth and rollout refresh. A 3x3…

science

Sources

• T1Prompt Breadth and Rollout Refresh Interact in On-Policy DistillationarXiv24 Sept, 04:00