English
Related papers

Related papers: Probing RLVR training instability through the lens…

200 papers

Large language models have achieved significant reasoning improvements through reinforcement learning with verifiable rewards (RLVR). Yet as model capabilities grow, constructing high-quality reward signals becomes increasingly difficult,…

Machine Learning · Computer Science 2026-04-21 Salman Rahman , Jingyan Shen , Anna Mordvina , Hamid Palangi , Saadia Gabriel , Pavel Izmailov

Alignment of large language models (LLMs) typically involves training a reward model on preference data, followed by policy optimization with respect to the reward model. However, optimizing policies with respect to a single reward model…

Machine Learning · Computer Science 2025-07-23 Debangshu Banerjee , Kintan Saha , Aditya Gopalan

Reinforcement learning with verifiable reward (RLVR) has been instrumental in eliciting strong reasoning capabilities from large language models (LLMs) via long chains of thought (CoT). During RLVR training, we formalize and systemically…

Computation and Language · Computer Science 2026-02-24 Cheonbok Park , Jeonghoon Kim , Joosung Lee , Sanghwan Bae , Jaegul Choo , Kang Min Yoo

Large Language Models (LLMs) are now widely used across many domains. With their rapid development, Reinforcement Learning with Verifiable Rewards (RLVR) has surged in recent months to enhance their reasoning and understanding abilities.…

Artificial Intelligence · Computer Science 2025-10-14 Jiecheng Zhou , Qinghao Hu , Yuyang Jin , Zerui Wang , Peng Sun , Yuzhe Gu , Wenwei Zhang , Mingshu Zhai , Xingcheng Zhang , Weiming Zhang

While Reinforcement Learning for Verifiable Rewards (RLVR) is powerful for training large reasoning models, its training dynamics harbor a critical challenge: RL overfitting, where models gain training rewards but lose generalization. Our…

Artificial Intelligence · Computer Science 2025-11-07 Zeng Zhiyuan , Jiashuo Liu , Zhangyue Yin , Ge Zhang , Wenhao Huang , Xipeng Qiu

Reinforcement Learning with Verifiable Rewards (RLVR) has recently strengthened LLM reasoning, but its focus on final answer correctness leaves a critical gap: it does not ensure the robustness of the reasoning process itself. We adopt a…

Machine Learning · Computer Science 2026-02-10 Hyunseok Lee , Soheil Abbasloo , Jihoon Tack , Jinwoo Shin

Reinforcement learning with verifiable rewards (RLVR) for Large Reasoning Models hinges on baseline estimation for variance reduction, but existing approaches pay a heavy price: PPO requires a policy-model scale critic, while GRPO needs…

Machine Learning · Computer Science 2026-05-12 Yunho Choi , Jongwon Lim , Woojin Ahn , Minjae Oh , Jeonghoon Shim , Yohan Jo

Reliable evaluation is fundamental to the progress of Large Language Models (LLMs), yet the evaluation process during pre-training is plagued by significant instability that obscures true learning dynamics. In this work, we systematically…

Computation and Language · Computer Science 2026-03-17 Jiapeng Wang , Changxin Tian , Kunlong Chen , Ziqi Liu , Jiaxin Mao , Wayne Xin Zhao , Zhiqiang Zhang , Jun Zhou

Reinforcement learning with verifiable rewards (RLVR) has advanced the reasoning capabilities of large language models. However, existing methods rely solely on outcome rewards, without explicitly optimizing verification or leveraging…

Software Engineering · Computer Science 2025-10-22 Yiyang Jin , Kunzhao Xu , Hang Li , Xueting Han , Yanmin Zhou , Cheng Li , Jing Bai

Reward hacking -- where RL agents exploit gaps in misspecified reward functions -- has been widely observed, but not yet systematically studied. To understand how reward hacking arises, we construct four RL environments with misspecified…

Machine Learning · Computer Science 2022-02-15 Alexander Pan , Kush Bhatia , Jacob Steinhardt

Stable pre-training is essential for achieving better-performing language models. However, tracking pre-training stability by calculating gradient variance at every step is impractical due to the significant computational costs. We explore…

Computation and Language · Computer Science 2024-09-13 Woojin Chung , Jiwoo Hong , Na Min An , James Thorne , Se-Young Yun

Existing reinforcement learning (RL)-based post-training methods for large language models have advanced rapidly, yet their design has largely been guided by heuristics rather than systematic theoretical principles. This gap limits our…

Machine Learning · Statistics 2026-01-16 Zixun Huang , Jiayi Sheng , Zeyu Zheng

As Large Reasoning Models (LRMs) are increasingly deployed, auditing their chain-of-thought (CoT) traces for safety becomes critical. Recent work has reported that monitorability--the degree to which CoT faithfully and informatively…

Artificial Intelligence · Computer Science 2026-02-05 Zidi Xiong , Shan Chen , Himabindu Lakkaraju

Reinforcement Learning with Verifiable Rewards (RLVR), which uses simple binary feedback to post-train large language models, has found significant empirical success. However, a principled understanding of why it works is lacking. This…

Machine Learning · Computer Science 2026-05-08 Joe Suk , Yaqi Duan

Reinforcement learning with verifiable rewards (RLVR) has become central to post-training reasoning models, yet a key limitation of existing studies is their narrow view of the reasoning space: difficulty is treated as reasoning depth…

Computation and Language · Computer Science 2026-05-27 Yihua Zhu , Qianying Liu , Fei Cheng , Jiaxin Wang , Akiko Aizawa , Sadao Kurohashi , Hidetoshi Shimodaira

Reinforcement learning with verifiable rewards (RLVR) has improved the reasoning ability of large language models, yet training remains costly because many rollouts contribute little to optimization, considering the amount of computation…

Machine Learning · Computer Science 2026-02-20 Yan Sun , Jia Guo , Stanley Kok , Zihao Wang , Zujie Wen , Zhiqiang Zhang

Reinforcement Learning with Verifiable Rewards (RLVR) serves as a cornerstone technique for enhancing the reasoning capabilities of Large Language Models (LLMs). However, its training is often plagued by \emph{entropy collapse}, a rapid…

Machine Learning · Computer Science 2026-04-30 Zhezheng Hao , Hong Wang , Haoyang Liu , Jian Luo , Jiarui Yu , Hande Dong , Qiang Lin , Can Wang , Jiawei Chen

Reinforcement Learning from Verifiable Rewards (RLVR) suffers from exploration inefficiency, where models struggle to generate successful rollouts, resulting in minimal learning signal. This challenge is particularly severe for tasks that…

Machine Learning · Computer Science 2026-03-20 Saaket Agashe , Jayanth Srinivasa , Gaowen Liu , Ramana Kompella , Xin Eric Wang

Learning systems are typically optimized by minimizing loss or maximizing reward, assuming that improvements in these signals reflect progress toward the true objective. However, when feedback reliability is unobservable, this assumption…

Machine Learning · Computer Science 2026-03-24 Zhipeng Zhang , Zhenjie Yao , Kai Li , Lei Yang

The ongoing debate on whether reinforcement learning with verifiable rewards (RLVR) expands or shrinks the reasoning capabilities of large language models (LLMs) remains unresolved. Some studies contend that RLVR mainly improves sampling…

Machine Learning · Computer Science 2025-10-07 Xinhao Yao , Lu Yu , Xiaolin Hu , Fengwei Teng , Qing Cui , Jun Zhou , Yong Liu