How RL Reshapes LLM Reasoning: Transferability, Coverage, and Scaling Laws

arXiv:2610.04158 · cs.LG, cs.AI, stat.ML · Submitted 2026-10-03 · Read on arXiv

cs.LG, cs.AI, stat.ML

Submitted: 2026-10-03

Updated: 2026-10-03

Terminology

• Scaling Laws — 1× in this paper · explained in 4 episodes
• Transferability — 1× in this paper · explained in 4 episodes

Sources

• The Coverage Principle: How Pre-Training Enables Post-Training
• Retaining by Doing: The Role of On-Policy Data in Mitigating Forgetting
• Reshaping Reasoning in LLMs: A Theoretical Analysis of RL Training Dynamics through Pattern Selection
• Revisiting Reinforcement Learning for LLM Reasoning from A Cross-Domain Perspective
• SFT Memorizes, RL Generalizes: A Comparative Study of Foundation Model Post-training
• Process Reinforcement through Implicit Rewards
• DeepSeek-R1: Incentivizing Reasoning Capability in LLMs via Reinforcement Learning
• SciKnowEval: Evaluating Multi-level Scientific Knowledge of Large Language Models
• Gemma 3 Technical Report
• Measuring Mathematical Problem Solving With the MATH Dataset
• Open-Reasoner-Zero: An Open Source Approach to Scaling Up Reinforcement Learning on the Base Model
• Does Math Reasoning Improve General LLM Capabilities? Understanding Transferability of LLM Reasoning
• LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code
• The Art of Scaling Reinforcement Learning Compute for LLMs
• ACPBench: Reasoning about Action, Change, and Planning
• Tulu 3: Pushing Frontiers in Open Language Model Post-Training
• Competition-Level Code Generation with AlphaCode
• ProRL: Prolonged Reinforcement Learning Expands Reasoning Boundaries in Large Language Models
• Understanding R1-Zero-Like Training: A Critical Perspective
• Reinforcement Learning with Verifiable Rewards: GRPO's Effective Loss, Dynamics, and Success Amplification

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: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks