Multi-Dimensional Behavioral Evaluation of Agentic Stock Prediction Systems Using Large Language Model Judges with Closed-Loop Reinforcement Learning Feedback

By Mohammad Al Ridhawi, Mahtab Haj Ali, Hussein Al Osman

Published 2026-05-16

Everscope rating

1137.9

Relevance to quantitative trading

9 / 10

Implementation complexity

8 / 10

Reproducibility

3 / 5

About this paper

Methodology: Multi-Dimensional Behavioral Evaluation with LLM Judges and Closed-Loop RL Feedback. Problem types: Time Series Forecasting, Reinforcement Learning, Anomaly Detection, Classification, Risk Management, Optimization.

arXiv:2605.05739 · Paper rankings

Open the interactive Everscope explorer for full analysis, charts, and paper battles.