Preprint · 2025 · Open access
Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning
arXiv (Cornell University) · Published 20 February 2025 · DOI 10.48550/arxiv.2502.14768
In brief
“Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning” is a 2025 preprint by Xie, Tian and 9 co-authors, published in arXiv (Cornell University). OpenAlex counts 365 citations as of 15 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).
- Citations
- 365
- Most cited in
- 2025 (204)
- Authors
- 10
Abstract
Inspired by the success of DeepSeek-R1, we explore the potential of rule-based reinforcement learning (RL) in large reasoning models. To analyze reasoning dynamics, we use synthetic logic puzzles as training data due to their controllable complexity and straightforward answer verification. We make some key technical contributions that lead to effective and stable RL training: a system prompt that emphasizes the thinking and answering process, a stringent format reward function that penalizes outputs for taking shortcuts, and a straightforward training recipe that achieves stable convergence. Our 7B model develops advanced reasoning skills-such as reflection, verification, and summarization-that are absent from the logic corpus. Remarkably, after training on just 5K logic problems, it demonstrates generalization abilities to the challenging math benchmarks AIME and AMC.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Tian, X., Gao, Z., Qingnan, R., Haoming, L., Yuqian, H., Bryan, D., Zhou, J. T., Qiu, K., Wu, Z., & Luo, C. (2025). Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2502.14768
MLA 9
Tian, Xie,, et al. “Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning.” arXiv (Cornell University), 2025. https://doi.org/10.48550/arxiv.2502.14768.
Chicago
Tian, Xie,, Zitian Gao, Ren, Qingnan, Luo, Haoming, Hong, Yuqian, Dai, Bryan, Joey Tianyi Zhou, Kai Qiu, Zhirong Wu, and Chong Luo. 2025. “Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2502.14768.
Harvard
Tian, X. et al. (2025) ‘Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning’, arXiv (Cornell University). doi:10.48550/arxiv.2502.14768.
IEEE
X. Tian et al., “Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning,” arXiv (Cornell University), 2025, doi: 10.48550/arxiv.2502.14768.
BibTeX
@misc{tian2025logic,
title = {{Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning}},
author = {Tian, Xie, and Gao, Zitian and Qingnan, Ren, and Haoming, Luo, and Yuqian, Hong, and Bryan, Dai, and Zhou, Joey Tianyi and Qiu, Kai and Wu, Zhirong and Luo, Chong},
howpublished = {arXiv (Cornell University)},
year = {2025},
doi = {10.48550/arxiv.2502.14768},
url = {https://doi.org/10.48550/arxiv.2502.14768}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Citations by year
| Year | Citations | Share of the best year |
|---|---|---|
| 2024 | 1 | |
| 2025 | 204 | |
| 2026 | 160 |
Topic and keywords
Physical Sciences › Computer Science › Artificial Intelligence › Topic Modeling
- large language models
- LLM reasoning
- logic puzzles
- reward function design
- training stability
- reflection
- answer verification
- summarization
Work with this paper
Most-cited papers in Computer Science
- ImageNet classification with deep convolutional neural networksAlex Krizhevsky, Ilya Sutskever, Geoffrey E. Hinton · Communications of the ACM · 2017 · 114,726 citations
- Long Short-Term MemorySepp Hochreiter, Jürgen Schmidhuber · Neural Computation · 1997 · 109,097 citations
- A Mathematical Theory of CommunicationClaude E. Shannon · Bell System Technical Journal · 1948 · 91,736 citations
- Deep learningYann LeCun, Yoshua Bengio, Geoffrey E. Hinton · Nature · 2015 · 90,226 citations
- Faster R-CNN: Towards Real-Time Object Detection with Region Proposal NetworksShaoqing Ren, Kaiming He, Ross Girshick and 1 more · IEEE Transactions on Pattern Analysis and Machine Intelligence · 2016 · 74,143 citations
Questions about this paper
How do I cite this paper in APA style?
Tian, X., Gao, Z., Qingnan, R., Haoming, L., Yuqian, H., Bryan, D., Zhou, J. T., Qiu, K., Wu, Z., & Luo, C. (2025). Logic-RL: Unleashing LLM Reasoning with Rule-Based Reinforcement Learning. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2502.14768
How many times has this paper been cited?
365 times, according to OpenAlex on 15 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access: read the free full text at arXiv (Cornell University).
Who wrote this paper?
Xie, Tian, Zitian Gao, Ren, Qingnan, Luo, Haoming, Hong, Yuqian, Dai, Bryan, Joey Tianyi Zhou, Kai Qiu, Zhirong Wu, Chong Luo.
What is the DOI of this paper?
10.48550/arxiv.2502.14768. The permanent link is https://doi.org/10.48550/arxiv.2502.14768.
Bibliographic data from OpenAlex (CC0), last checked 15 November 2026. ResearchGuru does not host this paper; the links go to the publisher or the repository that does.