FutureX-Pro: Extending Future Prediction to High-Value Vertical Domains
Abstract
Building upon FutureX, which established a live benchmark for general-purpose future prediction, this report introduces FutureX-Pro, including FutureX-Finance, FutureX-Retail, FutureX-PublicHealth, FutureX-NaturalDisaster, and FutureX-Search. These together form a specialized framework extending agentic future prediction to high-value vertical domains. While generalist agents demonstrate proficiency in open-domain search, their reliability in capital-intensive and safety-critical sectors remains under-explored. FutureX-Pro targets four economically and socially pivotal verticals: Finance, Retail, Public Health, and Natural Disaster. We benchmark agentic Large Language Models (LLMs) on entry-level yet foundational prediction tasks -- ranging from forecasting market indicators and supply chain demands to tracking epidemic trends and natural disasters. By adapting the contamination-free, live-evaluation pipeline of FutureX, we assess whether current State-of-the-Art (SOTA) agentic LLMs possess the domain grounding necessary for industrial deployment. Our findings reveal the performance gap between generalist reasoning and the precision required for high-value vertical applications.
Cite
@article{arxiv.2601.12259,
title = {FutureX-Pro: Extending Future Prediction to High-Value Vertical Domains},
author = {Jiashuo Liu and Siyuan Chen and Zaiyuan Wang and Zhiyuan Zeng and Jiacheng Guo and Liang Hu and Lingyue Yin and Suozhi Huang and Wenxin Hao and Yang Yang and Zerui Cheng and Zixin Yao and Lingyue Yin and Haoxin Liu and Jiayi Cheng and Yuzhen Li and Zezhong Ma and Bingjie Wang and Bingsen Qiu and Xiao Liu and Zeyang Zhang and Zijian Liu and Jinpeng Wang and Mingren Yin and Tianci He and Yali Liao and Yixiao Tian and Zhenwei Zhu and Anqi Dai and Ge Zhang and Jingkai Liu and Kaiyuan Zhang and Wenlong Wu and Xiang Gao and Xinjie Chen and Zhixin Yao and Zhoufutu Wen and B. Aditya Prakash and Jose Blanchet and Mengdi Wang and Nian Si and Wenhao Huang},
journal= {arXiv preprint arXiv:2601.12259},
year = {2026}
}
Comments
21 pages