AlphaSeek: Trajectory-Level Self-Iterative Factor Mining Framework for Multi-Source Financial Data

2026-08-13

AI Summary

This study addresses the subjectivity and lack of end-to-end feedback in quantitative factor mining by proposing a Large Language Model-driven automated factor generation framework. We introduce a novel trajectory-level evolutionary mechanism combined with self-iterative redundancy-aware ensemble, integrating multi-source information summarization and evolutionary operator search to achieve closed-loop optimization from hypothesis formulation to backtesting. Experiments on the CSI300 index demonstrate superior performance with an annualized return rate of 8.28%, an information ratio of 1.29, and an information coefficient of 0.0454. Furthermore, zero-shot evaluation on the CSI500 validates strong cross-market transferability. These results confirm that the proposed approach significantly enhances both the robustness and generalizability of automated factor discovery processes.

Abstract

With the rapid rise of large language models, LLM-driven quantitative factor mining has become an increasingly active research area. However, existing methods still suffer from subjective direction design, limited integration of up-to-date multi-source information, semantic drift, factor redundancy, and the absence of an end-to-end feedback loop from factor discovery to portfolio backtesting. To address these limitations, we propose AlphaSeek, an end-to-end factor mining framework for quantitative investment that integrates automated direction discovery, trajectory-level factor evolution mining and self-iterative portfolio optimization. AlphaSeek first collects and summarizes multi-source financial information to identify promising mining directions. It then performs trajectory-level factor mining by extending the optimization unit from a single factor expression to a complete research trajectory covering hypothesis generation, factor construction, validation, backtesting, and feedback. Based on this design, we introduce evolution operators - parallel direction expansion, mutation and crossover - to improve search diversity, refinement quality and factor robustness. Finally, AlphaSeek constructs a self-iterative factor portfolio, allowing newly discovered factors to interact with an existing state-of-the-art(SOTA) factor library under redundancy-aware constraints. Experiments on CSI300 show that AlphaSeek achieves the strongest overall strategy-level performance on CSI300 with ARR of 8.28%, IR of 1.29 and MDD of 6.28%, while remaining competitive on factor predictive metrics with IC of 0.0454, while factors mined on CSI300 also achieve strong time-series return performance on CSI500 than other models, suggesting promising cross-market transferability under a zero-shot setting.

Problem

• Quantitative Factor Mining
• Large Language Models
• Semantic Drift
• Factor Redundancy
• End-to-End Feedback Loop

Innovation

• Trajectory-Level Factor Mining
• Self-Iterative Optimization
• Evolution Operators
• End-to-End Framework
• Multi-Source Financial Data