lasgroup/SDPO

GitHub

Reinforcement Learning via Self-Distillation (SDPO)

Python981 stars111 forksApache-2.0

distillation llm reasoning rl

Momentum

lasgroup/SDPO has 981 GitHub stars.

It currently ranks #38422 daily, #38572 weekly, #39556 monthly by momentum on Trending Repos; its best-ever rank is #4545.

Rankings

PeriodRankScore

Daily

#384220.0

Weekly

#385720.0

Monthly

#395560.0

How scoring works

Engagement

Total stars

Daily new stars

Embeds

Best rank #4545

Format

Markdown

[![lasgroup/SDPO badge](https://trending-repos.com/badge/lasgroup/SDPO.svg?v=3)](https://trending-repos.com/repositories/lasgroup/SDPO?utm_source=github&utm_medium=badge)

lasgroup/SDPO

Momentum

Rankings

Engagement

Similar trending repositories

Embeds

lasgroup/SDPO

Momentum

Rankings

Engagement

Similar trending repositories

Embeds