Metraipolis › archive › dc_074fa25fdda7
OC-OPD: Observation-coordinated Supervision for On-Policy Distillation — arXiv:2611.01965
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2611.01965
- License (as archived)
- CC0-1.0
- Archived
- 2026-11-03T10:57:34.711Z
- SHA-256 of the archived text
65b615aeeba3f5aedfdbc444a05d807c172823e777cc3314725c673c9377452f- Stable URL of this page
https://metraipolis.com/documento/dc_074fa25fdda7
How to cite
OC-OPD: Observation-coordinated Supervision for On-Policy Distillation — arXiv:2611.01965 — arXiv (en), https://arxiv.org/abs/2611.01965 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_074fa25fdda7 (Metraipolis archive dc_074fa25fdda7, sha256 65b615aeeba3f5aedfdbc444a05d807c172823e777cc3314725c673c9377452f)
BibTeX
@misc{dc074fa25fdda7,
title = {OC-OPD: Observation-coordinated Supervision for On-Policy Distillation — arXiv:2611.01965},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2611.01965},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_074fa25fdda7; sha256 65b615aeeba3f5aedfdbc444a05d807c172823e777cc3314725c673c9377452f},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.01965. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
Reverse and forward Kullback--Leibler (KL) divergences provide complementary supervision in on-policy distillation: reverse KL refines the student's active token rankings, and forward KL allocates mass to teacher-supported candidates. At student-generated prefixes, we form a four-observation matrix by treating the student and teacher as both candidate sources and evaluators. The diagonal self-observations measure uncertainty, while source-centered cross-observations recover the two KL objectives. We establish an exact identity equating the Fisher inverse-metric coupling of their logit gradients with the sum of these contrasts, yielding a common local descent direction for distinct positive distributions. Observation-coordinated on-policy distillation (OC-OPD) uses observations to set preferences, closed-form bargaining to coordinate gradients, and distributional discrepancy to scale the update. Across three heterogeneous student settings (non-thinking generation, thinking generation, and base-model initialization), OC-OPD achieves strong gains in mathematical reasoning and transfers effectively to out-of-distribution scientific question answering and code generation. Our code is publicly available at: this https URL . Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.AI, nuovi lavori del giorno). Categorie arXiv: Machine Learning (cs.LG) ; Artificial Intelligence (cs.AI); Optimization and Control (math.OC). Pagina del lavoro: https://arxiv.org/abs/2611.01965 — autori e testo completo sono lì (qui omessi i nomi: nessun dato personale nel deposito). Licenza: titolo e abstract sono metadati arXiv distribuiti in CC0 1.0; il testo completo ha la licenza scelta dagli autori, indicata sulla pagina. Letto il 2026-11-03T10:49:38.606Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).