MENTOR: a multi-agent framework for event and narrative trend prediction with optimized reasoning
Special Feature on Theories and Applications of Financial Large Models|Updated:2025-11-18
|
MENTOR: a multi-agent framework for event and narrative trend prediction with optimized reasoning
Enhanced Publication
MENTOR:一种基于优化推理的事件与叙事趋势预测多智能体框架
“In the realm of narrative economics, a new multi-agent framework called MENTOR has been developed to predict event and narrative trends, offering a more reliable approach to forecasting financial market outcomes. Expert researchers have integrated teacher-student iterative reasoning with progressive subtasks, outperforming recent baselines in event prediction and industry ranking tasks, and demonstrating practical improvements in investment performance.”
Frontiers of Information Technology & Electronic EngineeringVol. 26, Issue 10, Pages: 1847-1861(2025)
Affiliations:
1.Tsinghua Shenzhen International Graduate School, Tsinghua University, Shenzhen 518055, China
2.E Fund Management Co., Ltd., Guangzhou 510620, China
3.Department of Computer Science, The University of Manchester, Manchester M139PL, UK
Liyuan CHEN, Gaoguo JIA, Dongsheng GU, et al. MENTOR: a multi-agent framework for event and narrative trend prediction with optimized reasoning[J]. Frontiers of Information Technology & Electronic Engineering, 2025, 26(10): 1847-1861.
DOI:
Liyuan CHEN, Gaoguo JIA, Dongsheng GU, et al. MENTOR: a multi-agent framework for event and narrative trend prediction with optimized reasoning[J]. Frontiers of Information Technology & Electronic Engineering, 2025, 26(10): 1847-1861. DOI: 10.1631/FITEE.2500608.
MENTOR: a multi-agent framework for event and narrative trend prediction with optimized reasoningEnhanced Publication