Researchers release paper on LineageEvo, a lineage-aware LLM evolutionary factor mining method
Get up to speed
Xintong Liu, Chenkai Zhang, Yanzhao Ma, and Peng Yang published a research paper proposing LineageEvo, a lineage-aware LLM evolutionary factor mining method. It optimizes the large language model-driven evolutionary factor mining process through two mechanisms: lineage-conditioned LLM mutation and lineage hierarchical search allocation. The method was evaluated on factor predictability, portfolio performance, search efficiency, and lineage and expression diversity in the CSI 500 and S&P 500 markets. The paper states that LineageEvo achieved the best predictive performance in both markets, while LSPS generated a more balanced lineage distribution and a factor pool with broader structure.
AI-generated from reports · 1 hr agoUpdated
Coverage timeline
Follow the coverage to understand different sides of the event.
- arXiv q-finEvolutionary Factor Mining for Large Language Models Based on Lineage-Aware Search Control
This paper proposes the LineageEvo method, which optimizes the LLM-driven evolutionary factor mining process through two mechanisms: lineage-conditioned LLM mutation and lineage hierarchical search allocation. The method was evaluated across factor predictability, portfolio performance, search efficiency, and diversity dimensions on the CSI 500 and S&P 500 markets, achieving optimal predictive performance in both markets.
Trend of this event
Not enough consecutive observations yet; trend is not drawn.