English
Related papers

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

200 papers

LLM-based search agents are increasingly trained on entity-centric synthetic data to solve complex, knowledge-intensive tasks. However, prevailing training methods like Group Relative Policy Optimization (GRPO) discard this rich entity…

Computation and Language · Computer Science 2026-02-25 Yida Zhao , Kuan Li , Xixi Wu , Liwen Zhang , Dingchu Zhang , Baixuan Li , Maojia Song , Zhuo Chen , Chenxi Wang , Xinyu Wang , Kewei Tu , Pengjun Xie , Jingren Zhou , Yong Jiang

Latent reasoning offers a more efficient alternative to explicit reasoning by compressing intermediate reasoning into continuous representations and substantially shortening reasoning chains. However, existing latent reasoning methods…

Machine Learning · Computer Science 2026-05-01 Jingcheng Deng , Zihao Wei , Liang Pang , Junhong Wu , Shicheng Xu , Zenghao Duan , Huawei Shen

We investigate a narrow but common failure mode of GRPO-style reinforcement learning in the context of sparse verifiable rewards: early updates contain more responses with negative advantages than those with positive advantages, while…

Machine Learning · Computer Science 2026-05-29 Mohamed Sana , Nicola Piovesan , Antonio De Domenico , Fadhel Ayed , Haozhe Zhang

Finetuning Large Vision-Language Models with reinforcement learning has emerged as a promising approach to enhance their capability in object-level grounding. However, existing methods, mainly based on GRPO, assign rewards at the response…

Computer Vision and Pattern Recognition · Computer Science 2026-05-18 Yuyuan Liu , Yiping Ji , Anjie Le , Jiayuan Zhu , Jiazhen Pan , Can Peng , Jiajun Deng , Fengbei Liu , Junde Wu

Large Language Models (LLMs) are increasingly deployed in business-critical domains such as finance, education, healthcare, and customer support, where users expect consistent and reliable recommendations. Yet LLMs often exhibit variability…

Machine Learning · Computer Science 2026-04-20 Sonal Prabhune , Balaji Padmanabhan , Kaushik Dutta

Group Relative Policy Optimization (GRPO) has emerged as a promising critic-free reinforcement learning paradigm for reasoning tasks. However, standard GRPO employs a coarse-grained credit assignment mechanism that propagates group-level…

Computation and Language · Computer Science 2026-01-13 Ziheng Li , Liu Kang , Feng Xiao , Luxi Xing , Qingyi Si , Zhuoran Li , Weikang Gong , Deqing Yang , Yanghua Xiao , Hongcheng Guo

By leveraging differentiable dynamics, Reparameterization Policy Gradient (RPG) achieves high sample efficiency. However, current approaches are hindered by two critical limitations: the under-utilization of computationally expensive…

Machine Learning · Computer Science 2026-02-09 Hai Zhong , Xun Wang , Zhuoran Li , Longbo Huang

Constrained decision-making is essential for designing safe policies in real-world control systems, yet simulated environments often fail to capture real-world adversities. We consider the problem of learning a policy that will maximize the…

Machine Learning · Computer Science 2026-02-10 Sourav Ganguly , Kishan Panaganti , Arnob Ghosh , Adam Wierman

In this work we introduce reinforcement learning techniques for solving lexicographic multi-objective problems. These are problems that involve multiple reward signals, and where the goal is to learn a policy that maximises the first reward…

Machine Learning · Computer Science 2022-12-29 Joar Skalse , Lewis Hammond , Charlie Griffin , Alessandro Abate

Numerous algorithms have been proposed to $\textit{align}$ language models to remove undesirable behaviors. However, the challenges associated with a very large state space and creating a proper reward function often result in various…

Computation and Language · Computer Science 2024-06-06 Suraj Anand , David Getzen

Balancing helpfulness and safety (harmlessness) is a critical challenge in aligning large language models (LLMs). Current approaches often decouple these two objectives, training separate preference models for helpfulness and safety, while…

Machine Learning · Computer Science 2025-02-28 Xiyue Peng , Hengquan Guo , Jiawei Zhang , Dongqing Zou , Ziyu Shao , Honghao Wei , Xin Liu

In Reinforcement Learning (abbreviated as RL), an agent interacts with the environment via a set of possible actions, and a reward is generated from some unknown distribution. The task here is to find an optimal set of actions such that the…

Machine Learning · Computer Science 2025-07-21 Aditi Anand , Suman Banerjee , Dildar Ali

While Retrieval-Augmented Generation (RAG) has exhibited promise in utilizing external knowledge, its generation process heavily depends on the quality and accuracy of the retrieved context. Large language models (LLMs) struggle to evaluate…

Computation and Language · Computer Science 2025-10-13 Shi-Qi Yan , Quan Liu , Zhen-Hua Ling

Search agents extend Large Language Models (LLMs) beyond static parametric knowledge by enabling access to up-to-date and long-tail information unavailable during pretraining. While reinforcement learning has been widely adopted for…

Machine Learning · Computer Science 2026-04-21 Junzhe Wang , Zhiheng Xi , Yajie Yang , Hao Luo , Shihan Dou , Tao Gui , Qi Zhang

Reinforcement learning is emerging as a primary driver for improving language model reasoning capabilities. A fundamental question is whether current reinforcement learning algorithms -- such as Group Relative Policy Optimization (GRPO),…

Machine Learning · Computer Science 2025-06-23 Andre He , Daniel Fried , Sean Welleck

In RL, given a prompt, we sample a group of completions from a model and score them. Two questions follow: which completions should gain probability mass, and how should the parameters move to realize that change? Standard policy-gradient…

Machine Learning · Computer Science 2026-04-08 Jean Kaddour

To facilitate efficient learning, policy gradient approaches to deep reinforcement learning (RL) are typically paired with variance reduction measures and strategies for making large but safe policy changes based on a batch of experiences.…

Machine Learning · Computer Science 2023-11-13 Jared Markowitz , Edward W. Staley

Reinforcement learning with verifiable rewards (RLVR), particularly Group Relative Policy Optimization (GRPO), has advanced LLM reasoning. However, GRPO suffers from three credit assignment failures: uniform token-level granularity that…

Machine Learning · Computer Science 2026-05-07 Song Yu , Li Li , Wenwen Zhao , Zhisheng Yang

Reinforcement learning (RL) is vital for optimizing large language models (LLMs). Recent Group Relative Policy Optimization (GRPO) estimates advantages using multiple on-policy outputs per prompt, leading to high computational costs and low…

Computation and Language · Computer Science 2025-06-12 Siheng Li , Zhanhui Zhou , Wai Lam , Chao Yang , Chaochao Lu

In neural combinatorial optimization (CO), reinforcement learning (RL) can turn a deep neural net into a fast, powerful heuristic solver of NP-hard problems. This approach has a great potential in practical applications because it allows…

Machine Learning · Computer Science 2021-07-14 Yeong-Dae Kwon , Jinho Choo , Byoungjip Kim , Iljoo Yoon , Youngjune Gwon , Seungjai Min