English
Related papers

Related papers: MO-GRPO: Mitigating Reward Hacking of Group Relati…

200 papers

Recent advances in reinforcement learning for foundation models, such as Group Relative Policy Optimization (GRPO), have significantly improved the performance of foundation models on reasoning tasks. Notably, the advantage function serves…

Artificial Intelligence · Computer Science 2025-09-26 Wenke Huang , Quan Zhang , Yiyang Fang , Jian Liang , Xuankun Rong , Huanjin Yao , Guancheng Wan , Ke Liang , Wenwen He , Mingjun Li , Leszek Rutkowski , Mang Ye , Bo Du , Dacheng Tao

Group Relative Policy Optimization (GRPO) enables stable and preference-oriented updates via group-wise comparisons for post-training video generation. However, GRPO directly optimizes reward-induced advantages. Under sustained…

Computer Vision and Pattern Recognition · Computer Science 2026-03-18 Rui Li , Yuanzhi Liang , Ziqi Ni , Haibing Huang , Chi Zhang , Xuelong Li

With the rapid advancement of large language models (LLMs), their robustness against adversarial manipulations, particularly jailbreak backdoor attacks, has become critically important. Existing approaches to embedding jailbreak…

Computation and Language · Computer Science 2025-09-25 Wence Ji , Jiancan Wu , Aiying Li , Shuyi Zhang , Junkang Wu , An Zhang , Xiang Wang , Xiangnan He

Engagement-optimized adaptive tutoring systems may prioritize short-term behavioral signals over sustained learning outcomes, creating structural incentives for reward hacking in reinforcement learning policies. We formalize this challenge…

Artificial Intelligence · Computer Science 2026-04-07 Oluseyi Olukola , Nick Rahimi

Group Relative Policy Optimization (GRPO), recently introduced by DeepSeek, is a critic-free reinforcement learning algorithm for fine-tuning large language models. GRPO replaces the value function in Proximal Policy Optimization (PPO) with…

Machine Learning · Computer Science 2026-03-24 Lei Pang , Jun Luo , Ruinan Jin

Policy optimization for large language models often suffers from sparse reward signals in multi-step reasoning tasks. Critic-free methods like GRPO assign a single normalized outcome reward to all tokens, providing limited guidance for…

Machine Learning · Computer Science 2026-02-04 Ruiyi Ding , Yongxuan Lv , Xianhui Meng , Jiahe Song , Chao Wang , Chen Jiang , Yuan Cheng

Recently, GRPO-based reinforcement learning has shown remarkable progress in optimizing flow-matching models, effectively improving their alignment with task-specific rewards. Within these frameworks, the policy update relies on…

Computer Vision and Pattern Recognition · Computer Science 2025-10-31 Jing Wang , Jiajun Liang , Jie Liu , Henglin Liu , Gongye Liu , Jun Zheng , Wanyuan Pang , Ao Ma , Zhenyu Xie , Xintao Wang , Meng Wang , Pengfei Wan , Xiaodan Liang

Solving tasks in Reinforcement Learning is no easy feat. As the goal of the agent is to maximize the accumulated reward, it often learns to exploit loopholes and misspecifications in the reward signal resulting in unwanted behavior. While…

Machine Learning · Computer Science 2018-12-27 Chen Tessler , Daniel J. Mankowitz , Shie Mannor

Group Relative Policy Optimization (GRPO) has recently shown strong performance in post-training large language models and vision-language models. It raises a question of whether the GRPO also significantly promotes the test-time adaptation…

Computer Vision and Pattern Recognition · Computer Science 2026-05-06 Yujun Li , Hongyuan Zhang , Yuan Yuan

Group Relative Policy Optimization (GRPO) has become the dominant method for reinforcement learning with verifiable rewards in large language models, but it suffers from two critical limitations: gradient vanishing and diversity collapse.…

Machine Learning · Computer Science 2026-05-20 Khiem Le , Phuc Nguyen , Youssef Mroueh , Chi-Heng Lin , Shangqian Gao , Ting Hua , Nitesh V. Chawla

The enhancement of reasoning capabilities in large language models (LLMs) has garnered significant attention, with supervised fine-tuning (SFT) and reinforcement learning emerging as dominant paradigms. While recent studies recognize the…

Artificial Intelligence · Computer Science 2026-03-17 Zhijie Wang

Recent progress in Large Language Model (LLM) reasoning is increasingly driven by the refinement of post-training loss functions and alignment strategies. However, standard Reinforcement Learning (RL) paradigms like Group Relative Policy…

Machine Learning · Computer Science 2026-01-28 Kishan Panaganti , Zhenwen Liang , Wenhao Yu , Haitao Mi , Dong Yu

Group Relative Policy Optimisation (GRPO) enhances large language models by estimating advantages across a group of sampled trajectories. However, mapping these trajectory-level advantages to policy updates requires aggregating token-level…

Since DeepSeek-R1 popularized, Group Relative Policy Optimization (GRPO) has become the core part of training Reasoning LLMs. However, we find some deficiency that influences RL stability and inference efficiency, like zero-variance in…

Computation and Language · Computer Science 2025-09-30 Chen Li , Nazhou Liu , Kai Yang

Reinforcement Learning with Verifiable Rewards (RLVR), primarily driven by the Group Relative Policy Optimization (GRPO) algorithm, is a leading approach for enhancing the reasoning abilities of Large Language Models (LLMs). Despite its…

Machine Learning · Computer Science 2025-10-21 Kangqi Ni , Zhen Tan , Zijie Liu , Pingzhi Li , Tianlong Chen

Group Relative Policy Optimization (GRPO) has emerged as an effective method for training reasoning models. While it computes advantages based on group mean, GRPO treats each output as an independent sample during the optimization and…

Artificial Intelligence · Computer Science 2026-03-16 Yu Li , Tian Lan , Zhengling Qi

Reinforcement learning from human feedback (RLHF) or verifiable rewards (RLVR), the standard paradigm for aligning LLMs or building recent SOTA reasoning models, is highly sensitive to noise from inconsistent or erroneous rewards. Yet, the…

Machine Learning · Computer Science 2026-05-20 Omar El Mansouri , Fathinah Asma Izzati , Mohamed El Amine Seddik , Salem Lahlou

Recent work shows that preference alignment objectives can be interpreted as divergence estimators between aligned (preferred) & unaligned (less-preferred) distributions, yielding a principled recipe for designing alignment losses. However,…

Machine Learning · Computer Science 2026-05-12 Rajdeep Haldar , Lantao Mei , Guang Lin , Yue Xing , Qifan Song

Most reinforcement learning algorithms seek a single optimal strategy that solves a given task. However, it can often be valuable to learn a diverse set of solutions, for instance, to make an agent's interaction with users more engaging, or…

Machine Learning · Computer Science 2024-01-09 Wentse Chen , Shiyu Huang , Yuan Chiang , Tim Pearce , Wei-Wei Tu , Ting Chen , Jun Zhu

The advantage function is a central concept in RL that helps reduce variance in policy gradient estimates. For language modeling, Group Relative Policy Optimization (GRPO) was proposed to use the within-group sample mean as a baseline for…

Machine Learning · Computer Science 2026-04-23 Hu Wang , Congbo Ma , Ian Reid , Mohammad Yaqub