arXiv Machine Learning June 5, 2026 By Mohammad Mahdi Salmani-Zarchi, Zahra Rahimi, Heshaam Faili, Mohammad Javad Dousti

MDP-GRPO: Stabilized Group Relative Policy Optimization for Multi-Constraint Instruction Following

Read the original on arXiv Machine Learning →

arXiv:2606. 06058v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards is ideal for multi-constraint instruction following, yet standard group-relative policy optimization (GRPO) becomes unstable under discrete, low-dispersion rewards, where within-group reward distributions are frequently homogeneous.

Summary generated by The Flow from the publisher's feed. The full article lives at arXiv Machine Learning.

arXiv Machine Learning

Jul 24

MO-GRPO: Mitigating Reward Hacking of Group Relative Policy Optimization on Multi-Objective Problems

arXiv:2509. 22047v3 Announce Type: replace Abstract: Group Relative Policy Optimization (GRPO) has been shown to be an effective algorithm when an accurate reward model is available.

By Yuki Ichihara, Yuu Jinnai, Tetsuro Morimura, Mitsuki Sakamoto, Ryota Mitsuhashi, Eiji Uchibe

nlp reinforcement-learning benchmarks

arXiv Machine Learning

Jun 5

SALT: When More Rollouts Don't Help in Group-Based Policy Optimization and How to Make Them Matter

arXiv:2606. 05800v1 Announce Type: new Abstract: Reinforcement learning with verifiable rewards (RLVR) often adopts GRPO-style group-relative updates, sampling multiple rollouts per prompt to construct normalized learning signals.

By Powei Chang, Jinpeng Zhang, Chaoqun Sun, MiniWell Tsao, Lianrui Li, Jianxiang Xiang, Chenyu Wang, Yukang Gao, Dongying Kong

reinforcement-learning benchmarks

arXiv AI

Jun 30

BV-Blend: Uncertainty-Weighted Historical Baselines for Stable Critic-Free RL with Verifiable Rewards

arXiv:2606. 28707v1 Announce Type: new Abstract: Critic-free reinforcement learning with verifiable rewards (RLVR), exemplified by Group Relative Policy Optimization (GRPO), avoids training a value function (critic) and reduces memory and compute overhead relative to critic-based PPO pipelines for aligning large language models.

By Yupeng Chang, Yuan Wu, Yi Chang

llms reinforcement-learning benchmarks

arXiv Machine Learning

Jun 9

Reformulate LLM Reinforcement Learning for Efficient Training under Black-box Discrepancy

arXiv:2606. 08779v1 Announce Type: new Abstract: Reinforcement Learning (RL) has emerged as a pivotal post-training paradigm, yet it frequently suffers from unpredictable sub-optimum performance or even training collapses.

By Jiashun Liu, Runze Liu, Xu Wan, Jing Liang, Hongyao Tang, Ling Pan

llms reinforcement-learning

arXiv AI

Jul 21

CIGPO: Contextual Information-Gain Policy Optimization for Multi-Turn Evidence-Reading LLM Agents

arXiv:2607. 16244v1 Announce Type: cross Abstract: Training multi-turn evidence-reading agents with outcome-only reinforcement learning is unstable because intermediate turns receive little direct credit.

By Hao Dou

llms agents reinforcement-learning

arXiv Machine Learning

Jun 30

On the Policy Gradient Foundations of Group Relative Policy Optimization: Credit Assignment, Gradient Sparsity, and Rank Collapse

arXiv:2606. 29238v1 Announce Type: new Abstract: Group Relative Policy Optimization (GRPO) eliminates the learned critic in PPO by using the mean reward of grouped rollouts as a baseline.

By Amritansh Mishra, Supriyo Chakraborty, Berkcan Kapusuzoglu

reinforcement-learning efficiency

Related stories

MO-GRPO: Mitigating Reward Hacking of Group Relative Policy Optimization on Multi-Objective Problems

SALT: When More Rollouts Don't Help in Group-Based Policy Optimization and How to Make Them Matter

BV-Blend: Uncertainty-Weighted Historical Baselines for Stable Critic-Free RL with Verifiable Rewards

Reformulate LLM Reinforcement Learning for Efficient Training under Black-box Discrepancy

CIGPO: Contextual Information-Gain Policy Optimization for Multi-Turn Evidence-Reading LLM Agents

On the Policy Gradient Foundations of Group Relative Policy Optimization: Credit Assignment, Gradient Sparsity, and Rank Collapse