English
Related papers

Related papers: Pass@K Policy Optimization: Solving Harder Reinfor…

200 papers

Current large language model post-training optimizes a risk-neutral objective that maximizes expected reward, yet evaluation relies heavily on risk-seeking metrics like Pass@k (at least one success in k trials) and Max@k (maximum reward…

Machine Learning · Computer Science 2025-08-05 Kaichen Zhang , Shenghao Gao , Yuzhong Hong , Haipeng Sun , Junwei Bao , Hongfei Jiang , Yang Song , Hong Dingqian , Hui Xiong

Repeated sampling with a verifier is the standard way to allocate test-time compute for code generation, with pass@$K$ as the canonical metric. Yet the standard policy class draws $K$ independent samples from a single answer distribution,…

Computation and Language · Computer Science 2026-05-27 Yilong Li , Suman Banerjee , Tong Che

Reinforcement learning with verifiable reward has recently emerged as a central paradigm for post-training large language models (LLMs); however, prevailing mean-based methods, such as Group Relative Policy Optimization (GRPO), suffer from…

Machine Learning · Computer Science 2025-10-02 Tao Ren , Jinyang Jiang , Hui Yang , Wan Tian , Minhao Zou , Guanghao Li , Zishi Zhang , Qinghao Wang , Shentao Qin , Yanjun Zhao , Rui Tao , Hui Shao , Yijie Peng

Reinforcement learning with verifiable rewards (RLVR) has advanced the reasoning capabilities of large language models (LLMs). However, prevailing RLVR methods exhibit a systematic bias toward exploitation over exploration, as evidenced by…

Artificial Intelligence · Computer Science 2025-10-22 Ruotian Peng , Yi Ren , Zhouliang Yu , Weiyang Liu , Yandong Wen

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

Reinforcement learning (RL) has emerged as a powerful tool for fine-tuning large language models (LLMs) to improve complex reasoning abilities. However, state-of-the-art policy optimization methods often suffer from high computational…

Machine Learning · Computer Science 2025-05-28 Kianté Brantley , Mingyu Chen , Zhaolin Gao , Jason D. Lee , Wen Sun , Wenhao Zhan , Xuezhou Zhang

Pass@k is a widely used performance metric for verifiable large language model tasks, including mathematical reasoning, code generation, and short-answer reasoning. It defines success if any of $k$ independently sampled solutions passes a…

Machine Learning · Computer Science 2026-02-27 Anas Barakat , Souradip Chakraborty , Khushbu Pahwa , Amrit Singh Bedi

Constrained reinforcement learning has achieved promising progress in safety-critical fields where both rewards and constraints are considered. However, constrained reinforcement learning methods face challenges in striking the right…

Machine Learning · Computer Science 2024-10-29 Jianmina Ma , Jingtian Ji , Yue Gao

Reinforcement learning (RL) has become a central paradigm for post-training large language models (LLMs), particularly for complex reasoning tasks, yet it often suffers from exploration collapse: policies prematurely concentrate on a small…

Machine Learning · Computer Science 2026-01-16 Zhiyuan Hu , Yucheng Wang , Yufei He , Jiaying Wu , Yilun Zhao , See-Kiong Ng , Cynthia Breazeal , Anh Tuan Luu , Hae Won Park , Bryan Hooi

Single-trajectory reinforcement learning (RL) methods aim to optimize policies from datasets consisting of (prompt, response, reward) triplets, where scalar rewards are directly available. This supervision format is highly practical, as it…

Machine Learning · Computer Science 2025-12-23 Bilal Faye , Hanane Azzag , Mustapha Lebbah

Reinforcement learning with verifiable rewards (RLVR), which typically adopts Pass@1 as the reward, has faced the issues in balancing exploration and exploitation, causing policies to prefer conservative actions, converging to a local…

Machine Learning · Computer Science 2025-08-15 Zhipeng Chen , Xiaobo Qin , Youbin Wu , Yue Ling , Qinghao Ye , Wayne Xin Zhao , Guang Shi

In the classical Reinforcement Learning from Human Feedback (RLHF) framework, Proximal Policy Optimization (PPO) is employed to learn from sparse, sentence-level rewards -- a challenging scenario in traditional deep reinforcement learning.…

Machine Learning · Computer Science 2025-05-22 Han Zhong , Zikang Shan , Guhao Feng , Wei Xiong , Xinle Cheng , Li Zhao , Di He , Jiang Bian , Liwei Wang

Reinforcement Learning (RL) has emerged as a powerful tool for neural combinatorial optimization, enabling models to learn heuristics that solve complex problems without requiring expert knowledge. Despite significant progress, existing RL…

Machine Learning · Computer Science 2025-05-14 Mingjun Pan , Guanquan Lin , You-Wei Luo , Bin Zhu , Zhien Dai , Lijun Sun , Chun Yuan

This note reconciles two seemingly distinct approaches to policy gradient optimization for the Pass@K objective in reinforcement learning with verifiable rewards: (1) direct REINFORCE-style methods, and (2) advantage-shaping techniques that…

Machine Learning · Computer Science 2026-03-24 Christos Thrampoulidis , Sadegh Mahdavi , Wenlong Deng

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

Classical reinforcement learning (RL) aims to optimize the expected cumulative rewards. In this work, we consider the RL setting where the goal is to optimize the quantile of the cumulative rewards. We parameterize the policy controlling…

Machine Learning · Computer Science 2022-02-17 Jinyang Jiang , Jiaqiao Hu , Yijie Peng

Heuristic algorithms such as simulated annealing, Concorde, and METIS are effective and widely used approaches to find solutions to combinatorial optimization problems. However, they are limited by the high sample complexity required to…

Machine Learning · Computer Science 2019-06-18 Qingpeng Cai , Will Hang , Azalia Mirhoseini , George Tucker , Jingtao Wang , Wei Wei

Constrained Reinforcement Learning (RL) aims to maximize the return while adhering to predefined constraint limits, which represent domain-specific safety requirements. In continuous control settings, where learning agents govern system…

Machine Learning · Computer Science 2025-09-12 Somnath Hazra , Pallab Dasgupta , Soumyajit Dey

Reinforcement Learning with Verifiable Rewards (RLVR) has achieved substantial gains in single-attempt accuracy (Pass@1) on reasoning tasks, yet often suffers from reduced multi-sample coverage (Pass@K), indicating diversity collapse. We…

Machine Learning · Computer Science 2026-05-04 Anamika Lochab , Bolian Li , Ruqi Zhang

Reinforcement Learning with Verifiable Rewards (RLVR) has proven effective for enhancing Large Language Models (LLMs) on complex reasoning tasks. However, existing methods suffer from an exploration dilemma: the sharply peaked initial…

Artificial Intelligence · Computer Science 2025-09-30 Yuhua Jiang , Jiawei Huang , Yufeng Yuan , Xin Mao , Yu Yue , Qianchuan Zhao , Lin Yan
‹ Prev 1 2 3 10 Next ›