Metraipolis › archive › dc_b98de367d5ba

When Teachers Mislead: Spurious-Signal-Aware On-Policy Distillation — arXiv:2608.03632

Source
arXiv (en)
Original
https://arxiv.org/abs/2608.03632
License (as archived)
CC0-1.0
Archived
2026-10-31T12:16:31.837Z
SHA-256 of the archived text
7cdfb634dc41336b3cd0f93caf399d2e87b059776b1434d35a2dedb3c68fed94
Stable URL of this page
https://metraipolis.com/documento/dc_b98de367d5ba

How to cite

When Teachers Mislead: Spurious-Signal-Aware On-Policy Distillation — arXiv:2608.03632 — arXiv (en), https://arxiv.org/abs/2608.03632 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_b98de367d5ba (Metraipolis archive dc_b98de367d5ba, sha256 7cdfb634dc41336b3cd0f93caf399d2e87b059776b1434d35a2dedb3c68fed94)
BibTeX
@misc{dcb98de367d5ba,
  title = {When Teachers Mislead: Spurious-Signal-Aware On-Policy Distillation — arXiv:2608.03632},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2608.03632},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_b98de367d5ba; sha256 7cdfb634dc41336b3cd0f93caf399d2e87b059776b1434d35a2dedb3c68fed94},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2608.03632. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

On-Policy distillation (OPD) transfers teacher capabilities by supervising student-sampled trajectories with dense token-level teacher signals. Recent selective OPD methods improve this process by prioritizing signals that are confident, informative, or learnable. However, the assumptions overlook a fundamental failure mode of language models: their token-level judgments can be driven by input-agnostic language priors, formatting conventions, or stereotyped reasoning templates rather than task-specific evidence. We refer to such optimization-relevant but weakly input-grounded supervision as spurious signals in OPD, which may produce large gradients while contributing little task-improving direction. To mitigate this issue, we propose SA-OPD, a Spurious-Signal-Aware On-Policy Distillation framework that identifies and filters misleading token-level supervision based on input-groundedness and optimization impact. SA-OPD introduces a lightweight input-groundedness proxy estimating whether a token-level distillation signal truly depends on the input. It then filters only tokens that simultaneously exhibit low input-groundedness and extreme distillation divergence, thereby removing high-impact spurious updates and achieving fine-grained OPD optimization. Extensive experiments on both large language model (LLM) and vision-language model (VLM) settings demonstrate that SA-OPD consistently outperforms Vanilla OPD and competitive selective methods. These results establish input-groundedness as a key dimension for OPD supervision selection and offer a simple, effective strategy for mitigating spurious updates.

Tema degli esploratori: agenti e intelligenza artificiale (retro-catalogo arXiv: annunci del 2026-08-05 in cs.AI).
Categorie arXiv: Artificial Intelligence (cs.AI).
Pagina del lavoro: https://arxiv.org/abs/2608.03632 — 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 2026-10-31T12:05:52.828Z dalla pagina https://arxiv.org/catchup/cs.AI/2026-08-05?abs=True (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).