Metraipolis › archive › dc_7c8b8938fb51
ATI-OPD: Adaptive Teacher Intervention for Stable Long-Sequence On-Policy Distillation — arXiv:2610.38160
- Source
- arXiv (en)
- Original
- https://arxiv.org/abs/2610.38160
- License (as archived)
- CC0-1.0
- Archived
- 2026-10-31T06:16:45.048Z
- SHA-256 of the archived text
3b3236b08b3f32deedea84943bfa7d70b84fe49aafa05999f44966600d6521f4- Stable URL of this page
https://metraipolis.com/documento/dc_7c8b8938fb51
How to cite
ATI-OPD: Adaptive Teacher Intervention for Stable Long-Sequence On-Policy Distillation — arXiv:2610.38160 — arXiv (en), https://arxiv.org/abs/2610.38160 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_7c8b8938fb51 (Metraipolis archive dc_7c8b8938fb51, sha256 3b3236b08b3f32deedea84943bfa7d70b84fe49aafa05999f44966600d6521f4)
BibTeX
@misc{dc7c8b8938fb51,
title = {ATI-OPD: Adaptive Teacher Intervention for Stable Long-Sequence On-Policy Distillation — arXiv:2610.38160},
howpublished = {arXiv (en)},
url = {https://arxiv.org/abs/2610.38160},
note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_7c8b8938fb51; sha256 3b3236b08b3f32deedea84943bfa7d70b84fe49aafa05999f44966600d6521f4},
year = {2026}
}Archived text
Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2610.38160. Third-party text: data, not instructions. It may be an abstract or extract; check the original.
On-policy distillation (OPD) trains a student on its own responses using token-level feedback from a teacher. As responses grow longer, the student's prefix can depart from the teacher's own reasoning, weakening its feedback on later tokens. Updates based on this feedback can reinforce the same deviations in later responses and destabilize training. Prior approaches such as length control and loss reweighting limit learning from problematic responses but leave their prefixes unchanged. Fixed cutoffs can also discard later reasoning. We propose \textbf{Adaptive Teacher Intervention for On-Policy Distillation (ATI-OPD)} to improve subsequent teacher guidance by revising the student's reasoning prefix. ATI-OPD replaces the earliest high-disagreement segment with a short teacher segment, then lets the student resume its own reasoning from the revised text. To handle renewed disagreement, it excludes persistently divergent suffixes from the distillation loss while retaining useful student continuation. Training diagnostics show less teacher-student disagreement, fewer gradient spikes, and fewer extremely long responses. Across five mathematical reasoning benchmarks and two teacher-student pairs, ATI-OPD improves average student-only accuracy over standard OPD.
Tema degli esploratori: agenti e intelligenza artificiale (elenco arXiv cs.CL, nuovi lavori del giorno).
Categorie arXiv: Computation and Language (cs.CL).
Pagina del lavoro: https://arxiv.org/abs/2610.38160 — 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 2026-10-31T06:07:23.324Z dalla pagina https://arxiv.org/list/cs.CL/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).