FinZero: Launching Multi-modal Financial Time Series Forecast with Large Reasoning Model

Fuente: arXiv
Guardado en:
Detalles Bibliográficos
Autores principales: Wang, Yanlong, Xu, Jian, Ma, Fei, Zhang, Hongkang, Yu, Hang, Gao, Tiantian, Wang, Yu, You, Haochen, Huang, Shao-Lun, Sun, Danny Dongning, Zhang, Xiao-Ping
Formato: Preprint
Publicado: 2025
Materias:
Acceso en línea:
Etiquetas: Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
_version_ 1866916945080614912
author Wang, Yanlong
Xu, Jian
Ma, Fei
Zhang, Hongkang
Yu, Hang
Gao, Tiantian
Wang, Yu
You, Haochen
Huang, Shao-Lun
Sun, Danny Dongning
Zhang, Xiao-Ping
author_facet Wang, Yanlong
Xu, Jian
Ma, Fei
Zhang, Hongkang
Yu, Hang
Gao, Tiantian
Wang, Yu
You, Haochen
Huang, Shao-Lun
Sun, Danny Dongning
Zhang, Xiao-Ping
contents Financial time series forecasting is both highly significant and challenging. Previous approaches typically standardized time series data before feeding it into forecasting models, but this encoding process inherently leads to a loss of important information. Moreover, past time series models generally require fixed numbers of variables or lookback window lengths, which further limits the scalability of time series forecasting. Besides, the interpretability and the uncertainty in forecasting remain areas requiring further research, as these factors directly impact the reliability and practical value of predictions. To address these issues, we first construct a diverse financial image-text dataset (FVLDB) and develop the Uncertainty-adjusted Group Relative Policy Optimization (UARPO) method to enable the model not only output predictions but also analyze the uncertainty of those predictions. We then proposed FinZero, a multimodal pre-trained model finetuned by UARPO to perform reasoning, prediction, and analytical understanding on the FVLDB financial time series. Extensive experiments validate that FinZero exhibits strong adaptability and scalability. After fine-tuning with UARPO, FinZero achieves an approximate 13.48\% improvement in prediction accuracy over GPT-4o in the high-confidence group, demonstrating the effectiveness of reinforcement learning fine-tuning in multimodal large model, including in financial time series forecasting tasks.
format Preprint
id arxiv_https___arxiv_org_abs_2509_08742
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle FinZero: Launching Multi-modal Financial Time Series Forecast with Large Reasoning Model
Wang, Yanlong
Xu, Jian
Ma, Fei
Zhang, Hongkang
Yu, Hang
Gao, Tiantian
Wang, Yu
You, Haochen
Huang, Shao-Lun
Sun, Danny Dongning
Zhang, Xiao-Ping
Computational Finance
Artificial Intelligence
Financial time series forecasting is both highly significant and challenging. Previous approaches typically standardized time series data before feeding it into forecasting models, but this encoding process inherently leads to a loss of important information. Moreover, past time series models generally require fixed numbers of variables or lookback window lengths, which further limits the scalability of time series forecasting. Besides, the interpretability and the uncertainty in forecasting remain areas requiring further research, as these factors directly impact the reliability and practical value of predictions. To address these issues, we first construct a diverse financial image-text dataset (FVLDB) and develop the Uncertainty-adjusted Group Relative Policy Optimization (UARPO) method to enable the model not only output predictions but also analyze the uncertainty of those predictions. We then proposed FinZero, a multimodal pre-trained model finetuned by UARPO to perform reasoning, prediction, and analytical understanding on the FVLDB financial time series. Extensive experiments validate that FinZero exhibits strong adaptability and scalability. After fine-tuning with UARPO, FinZero achieves an approximate 13.48\% improvement in prediction accuracy over GPT-4o in the high-confidence group, demonstrating the effectiveness of reinforcement learning fine-tuning in multimodal large model, including in financial time series forecasting tasks.
title FinZero: Launching Multi-modal Financial Time Series Forecast with Large Reasoning Model
topic Computational Finance
Artificial Intelligence
url https://arxiv.org/abs/2509.08742