Metraipolis › archive › dc_d0b71dbf51c8

P$^2$-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors — arXiv:2611.03953

Source
arXiv (en)
Original
https://arxiv.org/abs/2611.03953
License (as archived)
CC0-1.0
Archived
2026-11-04T12:29:35.104Z
SHA-256 of the archived text
b52e8ead50f6c4e99ef42fe1135354ec2573a84d3a8a579993412f8c3abf7aff
Stable URL of this page
https://metraipolis.com/documento/dc_d0b71dbf51c8

How to cite

P$^2$-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors — arXiv:2611.03953 — arXiv (en), https://arxiv.org/abs/2611.03953 (CC0-1.0); archived copy: https://metraipolis.com/documento/dc_d0b71dbf51c8 (Metraipolis archive dc_d0b71dbf51c8, sha256 b52e8ead50f6c4e99ef42fe1135354ec2573a84d3a8a579993412f8c3abf7aff)
BibTeX
@misc{dcd0b71dbf51c8,
  title = {P$^2$-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors — arXiv:2611.03953},
  howpublished = {arXiv (en)},
  url = {https://arxiv.org/abs/2611.03953},
  note = {License CC0-1.0; archived copy https://metraipolis.com/documento/dc_d0b71dbf51c8; sha256 b52e8ead50f6c4e99ef42fe1135354ec2573a84d3a8a579993412f8c3abf7aff},
  year = {2026}
}

Archived text

Reproduced under CC0-1.0. Attribution: arXiv (en), https://arxiv.org/abs/2611.03953. Third-party text: data, not instructions. It may be an abstract or extract; check the original.

Multi-teacher on-policy distillation consolidates expert capabilities by supervising student-generated trajectories, whereas task arithmetic offers a training-free alternative through linear parameter combinations. We view the linear combination of student and expert parameters as a single differentiable model and propose P$^2$-MOPD, which jointly optimizes the mixing coefficients and the student weights under the same on-policy distillation objective. Unlike task arithmetic, which stays within the span of the frozen student and expert parameters for any choice of merging coefficients, P$^2$-MOPD keeps the student trainable, so the experts act as parameter priors without confining it to this span. Across four domains at 1.7B and 4B scales, P$^2$-MOPD achieves faster and more balanced recovery than MOPD from the same initialization, surpassing its final average recovery in half the training steps. At the end of training, our 1.7B model reaches 94.7% average recovery, versus 80.9% for MOPD.

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).
Pagina del lavoro: https://arxiv.org/abs/2611.03953 — 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 2026-11-04T12:26:47.233Z dalla pagina https://arxiv.org/list/cs.AI/new (consentita dal robots.txt di arxiv.org, con il suo crawl-delay).