Conference paper · 2025 · Open access
Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning
neural information processing systems, 66634–66672 · Published 1 January 2025 · DOI 10.52202/085713-2002
In brief
“Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning” is a 2025 conference paper by Kongcheng Zhang and 7 co-authors, published in neural information processing systems. OpenAlex counts 40 citations as of 15 November 2026, which puts it in the top 2% of works published in 2025 by citations. It is open access: the full text is free to read at arXiv (Cornell University).
- Citations
- 40
- Most cited in
- 2026 (32)
- Field-weighted impact
- 14.14
- References
- 38
- Authors
- 8
Abstract
Recent advances of Reinforcement Learning (RL) have highlighted its potential in complex reasoning tasks, yet effective training often relies on external supervision, which limits the broader applicability. In this work, we propose a novel self-rewarding reinforcement learning framework to enhance Large Language Model (LLM) reasoning by leveraging the consistency of intermediate reasoning states across different reasoning trajectories. Our key insight is that correct responses often exhibit consistent trajectory patterns in terms of model likelihood: their intermediate reasoning states tend to converge toward their own final answers (high consistency) with minimal deviation toward other candidates (low volatility). Inspired by this observation, we introduce CoVo, an intrinsic reward mechanism that integrates Consistency and Volatility via a robust vector-space aggregation strategy, complemented by a curiosity bonus to promote diverse exploration. CoVo enables LLMs to perform RL in a self-rewarding manner, offering a scalable pathway for learning to reason without external supervision. Extensive experiments on diverse reasoning benchmarks show that CoVo achieves performance comparable to or even surpassing supervised RL. Our code is available at https://github.com/sastpg/CoVo.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Zhang, K., Yao, Q., Liu, S., Wang, Y., Lai, B., Ye, J., Song, M., & Tao, D. (2025). Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning. neural information processing systems, 66634–66672. https://doi.org/10.52202/085713-2002
MLA 9
Zhang, Kongcheng, et al. “Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning.” neural information processing systems, 2025, pp. 66634–66672. https://doi.org/10.52202/085713-2002.
Chicago
Zhang, Kongcheng, Qi Yao, Shunyu Liu, Yingjie Wang, Baisheng Lai, Jieping Ye, Mingli Song, and Dacheng Tao. 2025. “Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning.” neural information processing systems: 66634–66672. https://doi.org/10.52202/085713-2002.
Harvard
Zhang, K. et al. (2025) ‘Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning’, neural information processing systems, pp. 66634–66672. doi:10.52202/085713-2002.
IEEE
K. Zhang et al., “Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning,” neural information processing systems, pp. 66634–66672, 2025, doi: 10.52202/085713-2002.
BibTeX
@inproceedings{zhang2025consistent,
title = {{Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning}},
author = {Zhang, Kongcheng and Yao, Qi and Liu, Shunyu and Wang, Yingjie and Lai, Baisheng and Ye, Jieping and Song, Mingli and Tao, Dacheng},
booktitle = {neural information processing systems},
year = {2025},
pages = {66634--66672},
doi = {10.52202/085713-2002},
url = {https://doi.org/10.52202/085713-2002}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Computer Science › Artificial Intelligence › Reinforcement Learning in Robotics
- large language model reasoning
- intrinsic rewards
- reasoning paths
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?
Zhang, K., Yao, Q., Liu, S., Wang, Y., Lai, B., Ye, J., Song, M., & Tao, D. (2025). Consistent Paths Lead to Truth: Self-Rewarding Reinforcement Learning for LLM Reasoning. neural information processing systems, 66634–66672. https://doi.org/10.52202/085713-2002
How many times has this paper been cited?
40 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?
Kongcheng Zhang, Qi Yao, Shunyu Liu, Yingjie Wang, Baisheng Lai, Jieping Ye, Mingli Song, Dacheng Tao.
What is the DOI of this paper?
10.52202/085713-2002. The permanent link is https://doi.org/10.52202/085713-2002.
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.