Paper

When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning

Unreadunread

AI summary

arXiv:2605. 21606v1 Announce Type: new Abstract: On-policy self-distillation (OPSD) trains a student on its own rollouts using a privileged teacher, but its standard objective weights all generated tokens equally, implicitly treating the privileged teacher target as equally reliable at every student-visited prefix.

Reading status

Priority

About this paper

Authors: Xiaogeng Liu, Xinyan Wang, Yingzi Ma, Yechao Zhang, Chaowei Xiao
Released: 2026-05-23 (2mo ago)
Topic: Related to current project
Relation to my work: Read next because When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning overlaps with clean result "LoRA persona trained on <A> alone emits <B> at 23.5% when a co-trained partner learns <A>...<B>, vs 0% control on Qwen2.5-7B-Instruct (MODERATE confidence)", clean result "Leakage rate is a usable signal for recovering trigger-shaped phrases on Gaperon-1125-1B without knowing the hidden trigger itself (MODERATE confidence)", clean result "Language-mismatch LoRA SFT on Qwen2.5-7B leaks the trained completion language into bystander directives the model was never trained on, absent under same-language SFT (LOW confidence)". Matching terms: strong, latin, rect, under, correct, eval, prefix, token. Source: arxiv cs.LG (Machine Learning).
arXiv: 2605.21606

Ask Claude about this paper

Posts a comment with @claude so the runner spins up a Claude Code session scoped to When Are Teacher Tokens Reliable? Position-Weighted On-Policy Self-Distillation for Reasoning. Replies appear in the conversation below.

Comments

No comments yet.

Related records

No links yet.