Researchers from Princeton, Ant Group and Stanford Introduce AQuA: A Two-Part Agentic Framework for Autonomous Factor Discovery and Model Development in Quantitative Finance

Researchers from Princeton, Ant Group, and Stanford developed AQuA, a two-part AI framework that autonomously discovers quantitative finance factors and builds models while preventing methodological errors that cause overfitting. AQuA uses separate language-model-driven systems for crypto alpha factor discovery and US equities time-series modeling, employing an asymmetric freedom approach to maintain research integrity.