Preprint · 2025 · Open access
AgentRefine: Enhancing Agent Generalization through Refinement Tuning
arXiv (Cornell University) · Published 3 January 2025 · DOI 10.48550/arxiv.2501.01702
In brief
“AgentRefine: Enhancing Agent Generalization through Refinement Tuning” is a 2025 preprint by Dayuan Fu and 9 co-authors, published in arXiv (Cornell University). OpenAlex counts 45 citations as of 16 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).
- Citations
- 45
- Most cited in
- 2026 (23)
- References
- 23
- Authors
- 10
Abstract
Large Language Model (LLM) based agents have proved their ability to perform complex tasks like humans. However, there is still a large gap between open-sourced LLMs and commercial models like the GPT series. In this paper, we focus on improving the agent generalization capabilities of LLMs via instruction tuning. We first observe that the existing agent training corpus exhibits satisfactory results on held-in evaluation sets but fails to generalize to held-out sets. These agent-tuning works face severe formatting errors and are frequently stuck in the same mistake for a long while. We analyze that the poor generalization ability comes from overfitting to several manual agent environments and a lack of adaptation to new situations. They struggle with the wrong action steps and can not learn from the experience but just memorize existing observation-action relations. Inspired by the insight, we propose a novel AgentRefine framework for agent-tuning. The core idea is to enable the model to learn to correct its mistakes via observation in the trajectory. Specifically, we propose an agent synthesis framework to encompass a diverse array of environments and tasks and prompt a strong LLM to refine its error action according to the environment feedback. AgentRefine significantly outperforms state-of-the-art agent-tuning work in terms of generalization ability on diverse agent tasks. It also has better robustness facing perturbation and can generate diversified thought in inference. Our findings establish the correlation between agent generalization and self-refinement and provide a new paradigm for future research.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Fu, D., He, K., Wang, Y., Hong, W., Gongque, Z., Weihao, Z., Wei, W., Wang, J., Cai, X., & Xu, W. (2025). AgentRefine: Enhancing Agent Generalization through Refinement Tuning. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2501.01702
MLA 9
Fu, Dayuan, et al. “AgentRefine: Enhancing Agent Generalization through Refinement Tuning.” arXiv (Cornell University), 2025. https://doi.org/10.48550/arxiv.2501.01702.
Chicago
Fu, Dayuan, Keqing He, Yejie Wang, Wentao Hong, Zhuoma Gongque, Zeng, Weihao, Wang Wei, Jingang Wang, Xunliang Cai, and Weiran Xu. 2025. “AgentRefine: Enhancing Agent Generalization through Refinement Tuning.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2501.01702.
Harvard
Fu, D. et al. (2025) ‘AgentRefine: Enhancing Agent Generalization through Refinement Tuning’, arXiv (Cornell University). doi:10.48550/arxiv.2501.01702.
IEEE
D. Fu et al., “AgentRefine: Enhancing Agent Generalization through Refinement Tuning,” arXiv (Cornell University), 2025, doi: 10.48550/arxiv.2501.01702.
BibTeX
@misc{fu2025agentrefine,
title = {{AgentRefine: Enhancing Agent Generalization through Refinement Tuning}},
author = {Fu, Dayuan and He, Keqing and Wang, Yejie and Hong, Wentao and Gongque, Zhuoma and Weihao, Zeng, and Wei, Wang and Wang, Jingang and Cai, Xunliang and Xu, Weiran},
howpublished = {arXiv (Cornell University)},
year = {2025},
doi = {10.48550/arxiv.2501.01702},
url = {https://doi.org/10.48550/arxiv.2501.01702}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Topic and keywords
Physical Sciences › Computer Science › Artificial Intelligence › AI-based Problem Solving and Planning
- LLM agents
- self-refinement
- instruction tuning
- overfitting
- error correction
- open-source LLMs
- robustness to disturbances
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?
Fu, D., He, K., Wang, Y., Hong, W., Gongque, Z., Weihao, Z., Wei, W., Wang, J., Cai, X., & Xu, W. (2025). AgentRefine: Enhancing Agent Generalization through Refinement Tuning. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2501.01702
How many times has this paper been cited?
45 times, according to OpenAlex on 16 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?
Dayuan Fu, Keqing He, Yejie Wang, Wentao Hong, Zhuoma Gongque, Zeng, Weihao, Wang Wei, Jingang Wang, Xunliang Cai, Weiran Xu.
What is the DOI of this paper?
10.48550/arxiv.2501.01702. The permanent link is https://doi.org/10.48550/arxiv.2501.01702.
Bibliographic data from OpenAlex (CC0), last checked 16 November 2026. ResearchGuru does not host this paper; the links go to the publisher or the repository that does.