# Hugging Face Journal Club: Score Centering Stabilizes Off-policy Reinforcement Learning

**Published:** October 26, 2026
**Author:** Hugging Face
**Source:** YouTube — Hugging Face
**Language:** en-US
**HTML:** https://aiqazaqstan.com/en/articles/hugging-face-journal-club-score-centering-stabilizes-off-policy-reinforc

Yeah, the gist of the paper as you uh probably saw is just uh to identify the fundamental reason why RL learning for LLMs becomes unstable when there's a mismatch between the the inference engine and the training engine and uh the authors kind of uh go through the the math of…

> From the original source: **YouTube — Hugging Face**
> [Read full article at YouTube — Hugging Face](https://www.youtube.com/watch?v=mEtRe_KKYm0)

**Tags:** #Hugging Face #YouTube

---

*AI Qazaqstan — non-commercial informational portal about AI in Kazakhstan. Machine translations and summaries are AI-generated.*