Trajectory Learnability for Offline On-Policy Distillation with Imperfect Teachers
arXiv:2609.18321v2 Announce Type: replace-cross Abstract: Offline on-policy distillation gains efficiency by collecting student trajectories and teacher supervision once and reusing them throughout optimization. The same reuse makes imperfect…
Read the full story at arXiv cs.AI ↗
ImpactMinor 14/100
Why it mattersnimble: low (p 0.68)
RegionsGlobal
Published4 h ago (Fri, 23 Oct 2026 04:00:00 GMT)
RetrievedFri, 23 Oct 2026 04:00:25 GMT via rss
ClassifiedFri, 23 Oct 2026 04:06:58 GMT by nimble
AuthorYihao Ai, Beibei Lin, Weilong Yan, Huaijian Zhang