Revolve: Optimizing AI Systems by Tracking Response Evolution in Textual Optimization

Peiyan Zhang, Haibo Jin, Leyang Hu, Xinnuo Li, Liying Kang, Man Luo, Yangqiu Song, Haohan Wang
Proceedings of the 42nd International Conference on Machine Learning, PMLR 267:75216-75233, 2025.

Abstract

Recent advancements in large language models (LLMs) have significantly enhanced the ability of LLM-based systems to perform complex tasks through natural language processing and tool interaction. However, optimizing these LLM-based systems for specific tasks remains challenging, often requiring manual interventions like prompt engineering and hyperparameter tuning. Existing automatic optimization methods, such as textual feedback-based techniques (e.g., TextGrad), tend to focus on immediate feedback, analogous to using immediate derivatives in traditional numerical gradient descent. However, relying solely on such feedback can be limited when the adjustments made in response to this feedback are either too small or fluctuate irregularly, potentially slowing down or even stalling the optimization process. In this paper, we introduce $\textbf{REVOLVE}$, an optimization method that tracks how $\textbf{R}$esponses $\textbf{EVOLVE}$ across iterations in LLM systems. By focusing on the evolution of responses over time, REVOLVE enables more stable and effective optimization by making thoughtful, progressive adjustments at each step. Experiments across three tasks demonstrate the adaptability and efficiency of our proposal. Beyond its practical contributions, REVOLVE highlights a promising direction, where the rich knowledge from established optimization principles can be leveraged to enhance LLM systems, which paves the way for further advancements in this hybrid domain. Code is available at: https://llm-revolve.netlify.app.

Cite this Paper

BibTeX
@InProceedings{pmlr-v267-zhang25aj, title = {Revolve: Optimizing {AI} Systems by Tracking Response Evolution in Textual Optimization}, author = {Zhang, Peiyan and Jin, Haibo and Hu, Leyang and Li, Xinnuo and Kang, Liying and Luo, Man and Song, Yangqiu and Wang, Haohan}, booktitle = {Proceedings of the 42nd International Conference on Machine Learning}, pages = {75216--75233}, year = {2025}, editor = {Singh, Aarti and Fazel, Maryam and Hsu, Daniel and Lacoste-Julien, Simon and Berkenkamp, Felix and Maharaj, Tegan and Wagstaff, Kiri and Zhu, Jerry}, volume = {267}, series = {Proceedings of Machine Learning Research}, month = {13--19 Jul}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v267/main/assets/zhang25aj/zhang25aj.pdf}, url = {https://proceedings.mlr.press/v267/zhang25aj.html}, abstract = {Recent advancements in large language models (LLMs) have significantly enhanced the ability of LLM-based systems to perform complex tasks through natural language processing and tool interaction. However, optimizing these LLM-based systems for specific tasks remains challenging, often requiring manual interventions like prompt engineering and hyperparameter tuning. Existing automatic optimization methods, such as textual feedback-based techniques (e.g., TextGrad), tend to focus on immediate feedback, analogous to using immediate derivatives in traditional numerical gradient descent. However, relying solely on such feedback can be limited when the adjustments made in response to this feedback are either too small or fluctuate irregularly, potentially slowing down or even stalling the optimization process. In this paper, we introduce $\textbf{REVOLVE}$, an optimization method that tracks how $\textbf{R}$esponses $\textbf{EVOLVE}$ across iterations in LLM systems. By focusing on the evolution of responses over time, REVOLVE enables more stable and effective optimization by making thoughtful, progressive adjustments at each step. Experiments across three tasks demonstrate the adaptability and efficiency of our proposal. Beyond its practical contributions, REVOLVE highlights a promising direction, where the rich knowledge from established optimization principles can be leveraged to enhance LLM systems, which paves the way for further advancements in this hybrid domain. Code is available at: https://llm-revolve.netlify.app.} }
Endnote
%0 Conference Paper %T Revolve: Optimizing AI Systems by Tracking Response Evolution in Textual Optimization %A Peiyan Zhang %A Haibo Jin %A Leyang Hu %A Xinnuo Li %A Liying Kang %A Man Luo %A Yangqiu Song %A Haohan Wang %B Proceedings of the 42nd International Conference on Machine Learning %C Proceedings of Machine Learning Research %D 2025 %E Aarti Singh %E Maryam Fazel %E Daniel Hsu %E Simon Lacoste-Julien %E Felix Berkenkamp %E Tegan Maharaj %E Kiri Wagstaff %E Jerry Zhu %F pmlr-v267-zhang25aj %I PMLR %P 75216--75233 %U https://proceedings.mlr.press/v267/zhang25aj.html %V 267 %X Recent advancements in large language models (LLMs) have significantly enhanced the ability of LLM-based systems to perform complex tasks through natural language processing and tool interaction. However, optimizing these LLM-based systems for specific tasks remains challenging, often requiring manual interventions like prompt engineering and hyperparameter tuning. Existing automatic optimization methods, such as textual feedback-based techniques (e.g., TextGrad), tend to focus on immediate feedback, analogous to using immediate derivatives in traditional numerical gradient descent. However, relying solely on such feedback can be limited when the adjustments made in response to this feedback are either too small or fluctuate irregularly, potentially slowing down or even stalling the optimization process. In this paper, we introduce $\textbf{REVOLVE}$, an optimization method that tracks how $\textbf{R}$esponses $\textbf{EVOLVE}$ across iterations in LLM systems. By focusing on the evolution of responses over time, REVOLVE enables more stable and effective optimization by making thoughtful, progressive adjustments at each step. Experiments across three tasks demonstrate the adaptability and efficiency of our proposal. Beyond its practical contributions, REVOLVE highlights a promising direction, where the rich knowledge from established optimization principles can be leveraged to enhance LLM systems, which paves the way for further advancements in this hybrid domain. Code is available at: https://llm-revolve.netlify.app.
APA
Zhang, P., Jin, H., Hu, L., Li, X., Kang, L., Luo, M., Song, Y. & Wang, H.. (2025). Revolve: Optimizing AI Systems by Tracking Response Evolution in Textual Optimization. Proceedings of the 42nd International Conference on Machine Learning, in Proceedings of Machine Learning Research 267:75216-75233 Available from https://proceedings.mlr.press/v267/zhang25aj.html.

Related Material