On KL-Regularized Policy Optimization

cs.LG, cs.AI, cs.CL

Submitted: 2026-10-06

Updated: 2026-10-06

Comments: Project Page: https://github.com/yifanzhang-pro/KLPO

Code: https://github.com/yifanzhang-pro/KLPO

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

Sources

• Maximum a Posteriori Policy Optimisation
• REBEL: Reinforcement Learning via Regressing Relative Rewards
• IQ-Learn: Inverse soft-Q Learning for Imitation
• Offline Reinforcement Learning with Implicit Q-Learning
• Score Centering Stabilizes Off-policy Reinforcement Learning
• Advantage-Weighted Regression: Simple and Scalable Off-Policy Reinforcement Learning
• From $r$ to $Q^*$: Your Language Model is Secretly a Q-Function
• Proximal Policy Optimization Algorithms
• DeepSeekMath: Pushing the Limits of Mathematical Reasoning in Open Language Models
• Bellman Policy Optimization
• Self-Play Preference Optimization for Language Model Alignment
• Beyond Bradley-Terry Models: A General Preference Model for Language Model Alignment
• On the Design of KL-Regularized Policy Gradient Algorithms for LLM Reasoning

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