Preprint · 2025 · Open access
Don't Just Fine-tune the Agent, Tune the Environment
arXiv (Cornell University) · Published 11 October 2025 · DOI 10.48550/arxiv.2510.10197
In brief
“Don't Just Fine-tune the Agent, Tune the Environment” is a 2025 preprint by Siyuan Lu and 7 co-authors, published in arXiv (Cornell University). OpenAlex counts 15 citations as of 16 November 2026. It is open access: the full text is free to read at arXiv (Cornell University).
- Citations
- 15
- Most cited in
- 2026 (13)
- References
- 33
- Authors
- 8
Abstract
Large Language Model (LLM) agents show great promise for complex, multi-turn tool-use tasks, but their development is often hampered by the extreme scarcity of high-quality training data. Supervised fine-tuning (SFT) on synthetic data leads to overfitting, whereas standard reinforcement learning (RL) struggles with a critical cold-start problem and training instability. To address these challenges, we introduce $\textbf{Environment Tuning}$, a novel training paradigm that enables agents to learn complex behaviors directly from problem instances without relying on pre-collected expert trajectories. $\textbf{Environment Tuning}$ orchestrates this learning process through a structured curriculum, actionable environment augmentation that provides corrective feedback, and fine-grained progress rewards to ensure stable and efficient exploration. Using only 400 problem instances from Berkeley Function-Calling Leaderboard (BFCL) benchmark, our method not only achieves competitive in-distribution performance against strong baselines but also demonstrates superior out-of-distribution generalization, overcoming the performance collapse common to SFT-based approaches. Our work presents a paradigm shift from supervised fine-tuning on static trajectories to dynamic, environment-based exploration, paving the way for training more robust and data-efficient agents. The code is available at https://github.com/inclusionAI/AWorld-RL/tree/main/EnvTuning.
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Lu, S., Wang, Z., Zhang, H., Wu, Q., Gan, L., Zhuang, C., Gu, J., & Lin, T. (2025). Don't Just Fine-tune the Agent, Tune the Environment. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2510.10197
MLA 9
Lu, Siyuan, et al. “Don't Just Fine-tune the Agent, Tune the Environment.” arXiv (Cornell University), 2025. https://doi.org/10.48550/arxiv.2510.10197.
Chicago
Lu, Siyuan, Zilu Wang, Hongxuan Zhang, Qintong Wu, Leilei Gan, Chenyi Zhuang, Jinjie Gu, and Tao Lin. 2025. “Don't Just Fine-tune the Agent, Tune the Environment.” arXiv (Cornell University). https://doi.org/10.48550/arxiv.2510.10197.
Harvard
Lu, S. et al. (2025) ‘Don't Just Fine-tune the Agent, Tune the Environment’, arXiv (Cornell University). doi:10.48550/arxiv.2510.10197.
IEEE
S. Lu et al., “Don't Just Fine-tune the Agent, Tune the Environment,” arXiv (Cornell University), 2025, doi: 10.48550/arxiv.2510.10197.
BibTeX
@misc{lu2025just,
title = {{Don't Just Fine-tune the Agent, Tune the Environment}},
author = {Lu, Siyuan and Wang, Zilu and Zhang, Hongxuan and Wu, Qintong and Gan, Leilei and Zhuang, Chenyi and Gu, Jinjie and Lin, Tao},
howpublished = {arXiv (Cornell University)},
year = {2025},
doi = {10.48550/arxiv.2510.10197},
url = {https://doi.org/10.48550/arxiv.2510.10197}
}
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
- LLM agents
- function calling
- cold start
- reinforcement learning
- supervised fine-tuning
- curriculum learning
- out-of-distribution generalization
- data efficiency
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?
Lu, S., Wang, Z., Zhang, H., Wu, Q., Gan, L., Zhuang, C., Gu, J., & Lin, T. (2025). Don't Just Fine-tune the Agent, Tune the Environment. arXiv (Cornell University). https://doi.org/10.48550/arxiv.2510.10197
How many times has this paper been cited?
15 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?
Siyuan Lu, Zilu Wang, Hongxuan Zhang, Qintong Wu, Leilei Gan, Chenyi Zhuang, Jinjie Gu, Tao Lin.
What is the DOI of this paper?
10.48550/arxiv.2510.10197. The permanent link is https://doi.org/10.48550/arxiv.2510.10197.
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.