@inproceedings{wang-etal-2025-alpha,
    title = "Alpha-{GPT}: Human-{AI} Interactive Alpha Mining for Quantitative Investment",
    author = "Wang, Saizhuo  and
      Yuan, Hang  and
      Zhou, Leon  and
      Ni, Lionel  and
      Shum, Heung-Yeung  and
      Guo, Jian",
    editor = {Habernal, Ivan  and
      Schulam, Peter  and
      Tiedemann, J{\"o}rg},
    booktitle = "Proceedings of the 2025 Conference on Empirical Methods in Natural Language Processing: System Demonstrations",
    month = nov,
    year = "2025",
    address = "Suzhou, China",
    publisher = "Association for Computational Linguistics",
    url = "https://preview.aclanthology.org/author-page-yu-wang-polytechnic/2025.emnlp-demos.14/",
    doi = "10.18653/v1/2025.emnlp-demos.14",
    pages = "196--206",
    ISBN = "979-8-89176-334-0",
    abstract = "One of the most important tasks in quantitative investment research is mining new alphas (effective trading signals or factors). Traditional alpha mining methods, either hand-crafted factor synthesis or algorithmic factor mining (e.g., search with genetic programming), have inherent limitations, especially in implementing the ideas of quant researchers. In this work, we propose a new alpha mining paradigm by introducing human-AI interaction, and a novel prompt engineering algorithmic framework to implement this paradigm by leveraging the power of large language models. Moreover, we develop Alpha-GPT, a new interactive alpha mining system framework that provides a heuristic way to ``understand'' the ideas of quant researchers and outputs creative, insightful, and effective alphas. We demonstrate the effectiveness and advantage of Alpha-GPT via a number of alpha mining experiments. In particular, we evaluated Alpha-GPT{'}s performance in the \textbf{WorldQuant International Quant Championship}, where it demonstrated results comparable to those of top-performing human participants, ranking among \textbf{top-10} over 41000 teams worldwide. These findings suggest Alpha-GPT{'}s significant potential in generating highly effective alphas that may surpass human capabilities in quantitative investment strategies."
}