English
Related papers

Related papers: Tool Verification for Test-Time Reinforcement Lear…

200 papers

Effective interactive tool use requires agents to master Tool Integrated Reasoning (TIR): a complex process involving multi-turn planning and long-context dialogue management. To train agents for this dynamic process, particularly in…

Computation and Language · Computer Science 2025-09-19 Weiting Tan , Xinghua Qu , Ming Tu , Meng Ge , Andy T. Liu , Philipp Koehn , Lu Lu

Groundbreaking successes have been achieved by Deep Reinforcement Learning (DRL) in solving practical decision-making problems. Robotics, in particular, can involve high-cost hardware and human interactions. Hence, scrupulous evaluations of…

Artificial Intelligence · Computer Science 2020-10-20 Davide Corsi , Enrico Marchesini , Alessandro Farinelli

The successes of reinforcement learning in recent years are underpinned by the characterization of suitable reward functions. However, in settings where such rewards are non-intuitive, difficult to define, or otherwise error-prone in their…

Formal Languages and Automata Theory · Computer Science 2023-03-02 Mohammad Afzal , Sankalp Gambhir , Ashutosh Gupta , Krishna S , Ashutosh Trivedi , Alvaro Velasquez

Modern Large Language Models (LLMs) have shown rapid improvements in reasoning capabilities, driven largely by reinforcement learning (RL) with verifiable rewards. Here, we ask whether these LLMs can self-improve without the need for…

Computation and Language · Computer Science 2026-02-04 Yufan Zhuang , Chandan Singh , Liyuan Liu , Yelong Shen , Dinghuai Zhang , Jingbo Shang , Jianfeng Gao , Weizhu Chen

Test-time training (TTT) adapts model parameters on unlabeled test instances during inference time, which continuously extends capabilities beyond the reach of offline training. Despite initial gains, existing TTT methods for LRMs plateau…

Machine Learning · Computer Science 2026-04-22 Qingyang Zhang , Xinke Kong , Haitao Wu , Qinghua Hu , Minghao Wu , Baosong Yang , Yu Cheng , Yun Luo , Ganqu Cui , Changqing Zhang

Model-free reinforcement learning (RL) is a powerful, general tool for learning complex behaviors. However, its sample efficiency is often impractically large for solving challenging real-world problems, even with off-policy algorithms such…

Machine Learning · Computer Science 2020-02-25 Vitchyr Pong , Shixiang Gu , Murtaza Dalal , Sergey Levine

Learning from rewards (i.e., reinforcement learning or RL) and learning to imitate a teacher (i.e., teacher-student learning) are two established approaches for solving sequential decision-making problems. To combine the benefits of these…

Machine Learning · Computer Science 2024-02-21 Idan Shenfeld , Zhang-Wei Hong , Aviv Tamar , Pulkit Agrawal

Test-time scaling has been shown to substantially improve large language models' (LLMs) mathematical reasoning. However, for a large portion of mathematical corpora, especially theorem proving, RLVR's scalability is limited: intermediate…

Computation and Language · Computer Science 2025-11-24 Zhen Wang , Zhifeng Gao , Guolin Ke

Code verifiers play a critical role in post-verification for LLM-based code generation, yet existing supervised fine-tuning methods suffer from data scarcity, high failure rates, and poor inference efficiency. While reinforcement learning…

Artificial Intelligence · Computer Science 2026-02-02 Ji Shi , Peiming Guo , Meishan Zhang , Miao Zhang , Xuebo Liu , Min Zhang , Weili Guan

Learning to adapt pretrained language models to unlabeled, out-of-distribution data is a critical challenge, as models often falter on structurally novel reasoning tasks even while excelling within their training distribution. We introduce…

Computation and Language · Computer Science 2025-05-29 Mohammad Mahdi Moradi , Hossam Amer , Sudhir Mudur , Weiwei Zhang , Yang Liu , Walid Ahmed

In this work, we present Transitive Reinforcement Learning (TRL), a new value learning algorithm based on a divide-and-conquer paradigm. TRL is designed for offline goal-conditioned reinforcement learning (GCRL) problems, where the aim is…

Machine Learning · Computer Science 2026-02-24 Seohong Park , Aditya Oberai , Pranav Atreya , Sergey Levine

Reinforcement learning (RL) is a framework for solving sequential decision-making problems. In this work, we demonstrate that, surprisingly, RL emerges during the inference time of large language models (LLMs), a phenomenon we term…

Machine Learning · Computer Science 2026-04-28 Kefan Song , Amir Moeini , Peng Wang , Lei Gong , Rohan Chandra , Shangtong Zhang , Yanjun Qi

Large language models are increasingly used for complex reasoning tasks where high-quality offline data such as expert-annotated solutions and distilled reasoning traces are often available. However, in environments with sparse rewards,…

Artificial Intelligence · Computer Science 2025-08-11 Yihao Liu , Shuocheng Li , Lang Cao , Yuhang Xie , Mengyu Zhou , Haoyu Dong , Xiaojun Ma , Shi Han , Dongmei Zhang

Reinforcement learning with verifiable rewards (RLVR) has demonstrated promising potential to enhance the reasoning capabilities of large language models (LLMs) in domains such as mathematics and coding. However, its applications on…

Computation and Language · Computer Science 2026-05-19 Zhonghang Yuan , Zhefan Wang , Fang Hu , Zihong Chen , Jinzhe Li , Gang Li , Jie Ying , Huanjun Kong , Songyang Zhang , Nanqing Dong

The rapid progress of large language models (LLMs) has led to remarkable performance gains across a wide range of tasks. However, when handling long documents that exceed the model's context window limit, the entire context cannot be…

Computation and Language · Computer Science 2026-03-24 Li Wang , Yandong Wang , Xin Yu , Kui Zhang , Tianhao Peng , Wenjun Wu

Reinforcement Learning with Verifiable Rewards (RLVR) is widely used to improve reasoning in multiple domains, yet outcome-only scalar rewards are often sparse and uninformative, especially on failed samples, where they merely indicate…

Artificial Intelligence · Computer Science 2026-02-02 Xuancheng Li , Haitao Li , Yujia Zhou , YiqunLiu , Qingyao Ai

Continuous-time reinforcement learning (CTRL) provides a natural framework for sequential decision-making in dynamic environments where interactions evolve continuously over time. While CTRL has shown growing empirical success, its ability…

Machine Learning · Computer Science 2025-12-04 Runze Zhao , Yue Yu , Ruhan Wang , Chunfeng Huang , Dongruo Zhou

Trustworthy verifiers are essential for the success of reinforcement learning with verifiable reward (RLVR), which is the core methodology behind various large reasoning models such as DeepSeek-R1. In complex domains like mathematical…

Machine Learning · Computer Science 2025-10-08 Yuzhen Huang , Weihao Zeng , Xingshan Zeng , Qi Zhu , Junxian He

We introduce a control-tutored reinforcement learning (CTRL) algorithm. The idea is to enhance tabular learning algorithms by means of a control strategy with limited knowledge of the system model. By tutoring the learning process, the…

Optimization and Control · Mathematics 2022-04-14 Francesco De Lellis , Giovanni Russo , Mario di Bernardo

Reinforcement learning with verifiable rewards (RLVR) has demonstrated superior performance in enhancing the reasoning capability of large language models (LLMs). However, this accuracy-oriented learning paradigm often suffers from entropy…

Artificial Intelligence · Computer Science 2026-01-19 Hongye Cao , Zhixin Bai , Ziyue Peng , Boyan Wang , Tianpei Yang , Jing Huo , Yuyao Zhang , Yang Gao