← Back to all news
arXiv Machine Learning By Martin Marek, Max Ryabinin

Score Centering Stabilizes Off-policy Reinforcement Learning

Read the original on arXiv Machine Learning →

The paper introduces a method called score centering to address the training‑inference mismatch (TIM) that destabilizes reinforcement learning for large language models. By adding an additive correction term that cancels drift between training and inference engines, score centering stabilizes RL and can match or surpass importance‑sampling techniques, especially as model size and mismatch severity increase. The approach also composes with importance sampling, yielding further performance gains in staleness experiments.

Machine-generated by The Flow from the publisher's headline and feed description — not written or checked by a human. The full article lives at arXiv Machine Learning.

Related stories

Hugging Face Trending Papers
Jul 6

Turning Off-Policy Tokens On-Policy: A Plug-in Approach for Improving LLM Alignment

Reinforcement learning (RL) post-training for large language models (LLMs) follows a efficient paradigm of "rollout then update", which inevitably results in off-policy training data. To resolve this, Importance sampling (IS) is proposed, while the token-level ratios compound over long sequences, causing severe variance exploded.

llmsagentsreinforcement-learningbenchmarkssafety
More like this →
arXiv Machine Learning
Jun 30

The Mirage of Optimizing Training Policies: Monotonic Inference Policies as the Real Objective for LLM Reinforcement Learning

arXiv:2606. 29526v1 Announce Type: new Abstract: Reinforcement learning (RL) has gained growing attention in large language model (LLM) post-training, yet RL training remains fragile and can suffer from instability or collapse.

By Jing Liang, Hongyao Tang, Yi Ma, Yancheng He, Weixun Wang, Xiaoyang Li, Ju Huang, Wenbo Su, Jinyi Liu, Yan Zheng, Jianye Hao, Bo Zheng
llmsreinforcement-learning
More like this →
arXiv AI
Sep 11

BRACE: Anchored Bellman-Residual Correction for Stale Critics in Asynchronous RL

BRACE introduces an anchored Bellman‑residual correction to address stale critic bias in asynchronous reinforcement learning for language models. By limiting the correction horizon to a prefix of policy tokens and adding a constant‑weight Monte‑Carlo tail, it separates policy correction from reward propagation. The method improves mean@1 on BrowseComp‑Plus by 2.4% and runs 2.46× faster per step than synchronous training while staying stable 50 updates off‑policy.

By Guanqun Zhao, Zijun Xie, Binbin Zheng, Jiafeng Lu, Enlei Gong, Zeyu Chen
llmsagentsreinforcement-learningsafety
More like this →