Score Centering Stabilizes Off-policy Reinforcement Learning

arXiv:2609.20807 · cs.LG · Submitted 2026-09-17 · Read on arXiv

cs.LG

Submitted: 2026-09-17

Updated: 2026-09-17

Code: https://github.com/martin-marek/score-centering

License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/

Terminology

• Score Centering — 68× in this paper
• Reinforcement Learning — 19× in this paper · explained in 4 episodes
• Quantization — 16× in this paper · explained in 12 episodes
• Accuracy — 10× in this paper · explained in 1 episode
• Distillation — 7× in this paper · explained in 2 episodes
• Core — 4× in this paper · explained in 2 episodes
• KV cache — 3× in this paper · explained in 3 episodes
• Truncated Importance Sampling — 3× in this paper
• Ablation — 2× in this paper · explained in 1 episode
• Fine-Tuning — 2× in this paper · explained in 5 episodes
• Reinforcement Learning (RL) — 2× in this paper · explained in 25 episodes
• Supervised Fine-Tuning — 2× in this paper
• Action Space — 1× in this paper · explained in 1 episode
• Adam — 1× in this paper · explained in 2 episodes
• Agentic Coding — 1× in this paper · explained in 1 episode
• GRPO — 1× in this paper · explained in 6 episodes
• Generative Model — 1× in this paper · explained in 2 episodes
• Generative Models — 1× in this paper · explained in 2 episodes
• Loss Function — 1× in this paper · explained in 2 episodes
• Masked Importance Sampling — 1× in this paper
• Online Reinforcement Learning — 1× in this paper · explained in 2 episodes
• Supervised Fine-Tuning (SFT) — 1× in this paper · explained in 8 episodes
• Systematic Bias — 1× in this paper · explained in 3 episodes
• Transformer — 1× in this paper · explained in 2 episodes

Sources

• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• On the Effect of Negative Gradient in Group Relative Deep Reinforcement Optimization
• AReaL: A Large-Scale Asynchronous Reinforcement Learning System for Language Reasoning
• Stream of Search (SoS): Learning to Search in Language
• Distilling the Knowledge in a Neural Network
• Single-Rollout Asynchronous Optimization for Agentic Reinforcement Learning
• Tapered Off-Policy REINFORCE: Stable and efficient reinforcement learning for LLMs
• Every Step Evolves: Scaling Reinforcement Learning for Trillion-Scale Thinking Model
• Understanding R1-Zero-Like Training: A Critical Perspective
• Stabilizing MoE Reinforcement Learning by Aligning Training and Inference Routers
• MiniMax-M1: Scaling Test-Time Compute Efficiently with Lightning Attention
• OpenAI o1 System Card
• PipelineRL: Faster On-policy Reinforcement Learning for Long Sequence Generation
• INTELLECT-2: A Reasoning Model Trained Through Globally Decentralized Reinforcement Learning
• Defeating the Training-Inference Mismatch via FP16
• Rethinking the Trust Region in LLM Reinforcement Learning
• FP8-RL: A Practical and Stable Low-Precision Stack for LLM Reinforcement Learning
• Proximal Policy Optimization Algorithms
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• HybridFlow: A Flexible and Efficient RLHF Framework

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: Local-Consistency Optimisation for Bayesian Neural Networks
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks