Trendoday

P^2-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors

Multi-teacher on-policy distillation consolidates expert capabilities by supervising student-generated trajectories, whereas task arithmetic offers a training-free alternative through linear parameter

  • Upvotes9
–Rank now
#3Peak rank
2Days on list
381Score · 7d
Nov 9, 2026First seen
Rank · 7 days

About this paper

P$^2$-MOPD jointly optimizes mixing coefficients and student weights under an on-policy distillation objective, allowing experts to act as parameter priors and achieving faster, more balanced recovery than MOPD.

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…

multi-teacher on-policy distillationstudent-generated trajectoriestask arithmeticlinear parameter combinationsdifferentiable modelP$^2$-MOPDmixing coefficientsstudent weightson-policy distillation objectivefrozen student and expert parameters
Upvotes
9
Authors
11
Published
Nov 2, 2026

Summary and abstract excerpt via Hugging Face Papers ↗ · arXiv.

Trending history

P^2-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors was first seen on Hugging Face · Daily papers on Nov 9, 2026. Since then it has appeared in 18 checks on 2 different days. Its best position was #3 of 30, reached on Nov 10, 01:28 UTC. It dropped off the list after Nov 10, 17:28 UTC.

Rank history

Position on Hugging Face · Daily papers at every check (lower is better).

#1#11#20#30 07:2815:5800:2808:5817:28 Nov 9, 07:28 UTC · #24Nov 9, 09:28 UTC · #20Nov 9, 11:28 UTC · #9Nov 9, 13:28 UTC · #9Nov 9, 15:28 UTC · #6Nov 9, 17:28 UTC · #5Nov 9, 19:28 UTC · #5Nov 9, 21:28 UTC · #4Nov 9, 23:28 UTC · #4Nov 10, 01:28 UTC · #3Nov 10, 03:28 UTC · #3Nov 10, 05:28 UTC · #3Nov 10, 07:28 UTC · #3Nov 10, 09:28 UTC · #3Nov 10, 11:28 UTC · #3Nov 10, 13:28 UTC · #3Nov 10, 15:28 UTC · #4Nov 10, 17:28 UTC · #3

Last 14 days

Best rank per day (UTC).

Nov 3today

Metric history

Platform metric as reported by Hugging Face.

0163249 07:2815:5800:2808:5817:28 Nov 9, 07:28 UTC · 9Nov 9, 09:28 UTC · 10Nov 9, 11:28 UTC · 29Nov 9, 13:28 UTC · 34Nov 9, 15:28 UTC · 38Nov 9, 17:28 UTC · 39Nov 9, 19:28 UTC · 40Nov 9, 21:28 UTC · 42Nov 9, 23:28 UTC · 44Nov 10, 01:28 UTC · 44Nov 10, 03:28 UTC · 45Nov 10, 05:28 UTC · 46Nov 10, 07:28 UTC · 48Nov 10, 09:28 UTC · 49Nov 10, 11:28 UTC · 49Nov 10, 13:28 UTC · 49Nov 10, 15:28 UTC · 49Nov 10, 17:28 UTC · 49

Questions

Is P^2-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors trending on HF Daily Papers right now?

Not at the moment. It was last on the list on Nov 10, 17:28 UTC, after 2 days of trending.

What is the highest rank P^2-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors reached?

#3 on Hugging Face · Daily papers, on Nov 10, 01:28 UTC. Trendoday checks this list every 120 minutes.

How long has P^2-MOPD: Accelerating Multi-Teacher On-Policy Distillation via Parameter Priors been trending?

It was first seen on Nov 9, 2026 and has been on the list on 2 days (18 checks) since.