🛰️ Daily AI Frontier
‹ back to 2026-08-13

Towards Understanding On-Policy Distillation through the Lens of Test-Time Scaling

Research LLMs & Foundation Models

Ranking

Overall 77
Content 95
Popularity 34

Observed public metrics from 1 member.

Representative image for Towards Understanding On-Policy Distillation through the Lens of Test-Time Scaling

Merged summary

TL;DR - This study finds that on-policy distillation improves LLM reasoning mainly by making successful outputs more likely with small sampling budgets, rather than expanding the model’s ultimate reasoning capabilities.

  • OPD models sustain better avg@K across sampling budgets.
  • As K increases, pre-OPD models gradually overtake OPD models on pass@K.
  • Training shifts performance toward stronger small-K results at the expense of the large-K capability boundary.
  • At pass@1024, OPD makes more previously solvable problems unsolvable than it makes previously unsolvable problems solvable.

Sources (1)

Towards Understanding On-Policy Distillation through the Lens of Test-Time Scaling

arXiv cs.LG Xinmu Ge, Zizhuo Zhang, Yu Huang, Jianing Zhu, Lin Yuan, Wanli Gu, Weichang Wu, Weiran Huang, Xiaolu Zhang, Bo Han, Jun Zhou, Jiangchao Yao 2026-08-12 arXiv:2608.11829
Public signals Semantic Scholar citations 0 · Semantic Scholar influential citations 0
Providers: Hugging Face · N/A OpenAlex · N/A Publisher · N/A Semantic Scholar · Citations 0 · Influential citations 0 X · N/A Fetched 2026-09-12 14:27:38.219764 UTC

TL;DR - This study finds that on-policy distillation improves LLM reasoning mainly by making successful outputs more likely with small sampling budgets, rather than expanding the model’s ultimate reasoning capabilities.

  • OPD models sustain better avg@K across sampling budgets.
  • As K increases, pre-OPD models gradually overtake OPD models on pass@K.
  • Training shifts performance toward stronger small-K results at the expense of the large-K capability boundary.
  • At pass@1024, OPD makes more previously solvable problems unsolvable than it makes previously unsolvable problems solvable.
item →