English
Related papers

Related papers: OBLR-PO: A Theoretical Framework for Stable Reinfo…

200 papers

Reinforcement Learning with Verifiable Rewards (RLVR) has become a promising paradigm for scaling reasoning capabilities of Large Language Models (LLMs). However, the sparsity of binary verifier rewards often leads to low efficiency and…

Machine Learning · Computer Science 2026-05-22 Shuo Yang , Jinda Lu , Kexin Huang , Chiyu Ma , Shaohang Wei , Yuyang Liu , Guoyin Wang , Jingren Zhou , Li Yuan

Many reinforcement learning algorithms, particularly those that rely on return estimates for policy improvement, can suffer from poor sample efficiency and training instability due to high-variance return estimates. In this paper we…

Machine Learning · Computer Science 2026-01-06 Alexander W. Goodall , Edwin Hamel-De le Court , Francesco Belardinelli

The alignment of Large Language Models (LLMs) utilizes Reinforcement Learning from AI Feedback (RLAIF) for non-verifiable domains such as long-form question answering and open-ended instruction following. These domains often rely on LLM…

Machine Learning · Computer Science 2026-05-18 Nirmal Patel , Fei Wang , Inderjit S. Dhillon

A key challenge in applying reinforcement learning (RL) to diffusion large language models (dLLMs) lies in the intractability of their likelihood functions, which are essential for the RL objective, necessitating corresponding approximation…

Machine Learning · Computer Science 2025-10-15 Nianyi Lin , Jiajie Zhang , Lei Hou , Juanzi Li

Reinforcement learning substantially improves pretrained language models, but it remains understudied why critic-free methods such as PPO and GRPO work as well as they do, and when they should provide the largest gains. We develop a…

Machine Learning · Computer Science 2026-05-22 Arip Asadulaev , Daniil Ognev , Karim Salta , Martin Takac

While Reinforcement Learning (RL) shows promise in training tool-use Large Language Models (LLMs) using verifiable outcome rewards, existing methods largely overlook the potential of reasoning rewards based on chain-of-thought quality for…

Computation and Language · Computer Science 2026-01-16 Zihan Lin , Xiaohan Wang , Hexiong Yang , Jiajun Chai , Jie Cao , Guojun Yin , Wei Lin , Ran He

In offline reinforcement learning (RL), we seek to utilize offline data to evaluate (or learn) policies in scenarios where the data are collected from a distribution that substantially differs from that of the target policy to be evaluated.…

Machine Learning · Computer Science 2021-03-09 Ruosong Wang , Yifan Wu , Ruslan Salakhutdinov , Sham M. Kakade

We study the problem of training and fine-tuning expressive policies with online reinforcement learning (RL) given an offline dataset. Training expressive policy classes with online RL present a unique challenge of stable value…

Machine Learning · Computer Science 2026-05-01 Perry Dong , Qiyang Li , Dorsa Sadigh , Chelsea Finn

Recent advances in reinforcement learning with verifiable rewards (RLVR) show that large language models enhance their reasoning abilities when trained with verifiable signals. However, due to reward sparsity, effectiveness depends heavily…

Computation and Language · Computer Science 2026-01-27 Sanghwan Bae , Jiwoo Hong , Min Young Lee , Hanbyul Kim , JeongYeon Nam , Donghyun Kwak

Training stability remains a critical bottleneck for Group Relative Policy Optimization (GRPO), often manifesting as a trade-off between reasoning plasticity and general capability retention. We identify a root cause as the geometric…

Machine Learning · Computer Science 2026-02-09 Wenwen Qiang , Ziyin Gu , Jiahuan Zhou , Jie Hu , Jingyao Wang , Changwen Zheng , Hui Xiong

Large language model (LLM) alignment via reinforcement learning from human preferences (RLHF) suffers from unstable policy updates, ambiguous gradient directions, poor interpretability, and high gradient variance in mainstream pairwise…

Machine Learning · Computer Science 2026-05-12 Hao Yu

Model-based reinforcement learning (RL) is considered to be a promising approach to reduce the sample complexity that hinders model-free RL. However, the theoretical understanding of such methods has been rather limited. This paper…

Machine Learning · Computer Science 2021-02-16 Yuping Luo , Huazhe Xu , Yuanzhi Li , Yuandong Tian , Trevor Darrell , Tengyu Ma

Reinforcement learning (RL) post-training has increasingly demonstrated strong ability to elicit reasoning behaviors in large language models (LLMs). For training efficiency, rollouts are typically generated in an off-policy manner using an…

Artificial Intelligence · Computer Science 2026-02-02 Shiye Lei , Zhihao Cheng , Dacheng Tao

Policy-based methods have achieved remarkable success in solving challenging reinforcement learning problems. Among these methods, off-policy policy gradient methods are particularly important due to that they can benefit from off-policy…

Machine Learning · Computer Science 2024-05-07 Wenjia Meng , Qian Zheng , Long Yang , Yilong Yin , Gang Pan

Reinforcement learning (RL) post-training is crucial for LLM alignment and reasoning, but existing policy-based methods, such as PPO and DPO, can fall short of fixing shortcuts inherited from pre-training. In this work, we introduce…

Machine Learning · Computer Science 2025-10-21 Jin Peng Zhou , Kaiwen Wang , Jonathan Chang , Zhaolin Gao , Nathan Kallus , Kilian Q. Weinberger , Kianté Brantley , Wen Sun

We consider an improper reinforcement learning setting where a learner is given $M$ base controllers for an unknown Markov decision process, and wishes to combine them optimally to produce a potentially new controller that can outperform…

Machine Learning · Computer Science 2021-07-06 Mohammadi Zaki , Avinash Mohan , Aditya Gopalan , Shie Mannor

Reinforcement Learning (RL) has emerged as a transformative approach for aligning and enhancing Large Language Models (LLMs), addressing critical challenges in instruction following, ethical alignment, and reasoning capabilities. This…

Artificial Intelligence · Computer Science 2025-07-08 Saksham Sahai Srivastava , Vaneet Aggarwal

Reinforcement learning (RL) algorithms such as PPO and GRPO are widely used to train large language models (LLMs) for multi-turn agentic tasks. However, in off-policy training pipelines, these methods often exhibit unstable optimization…

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…

We study the problem of off-policy evaluation (OPE) in Reinforcement Learning (RL), where the aim is to estimate the performance of a new policy given historical data that may have been generated by a different policy, or policies. In…

Machine Learning · Computer Science 2019-12-16 Aurélien F. Bibaut , Ivana Malenica , Nikos Vlassis , Mark J. van der Laan