ReSGA: A Large Tail Risk Model for Learning Value-at-Risk and Expected Shortfall

By Yichi Zhang, Ke Zhu, Zhoufan Zhu

Published 2026-06-03

Everscope rating

1636

Relevance to quantitative trading

9 / 10

Implementation complexity

8 / 10

Reproducibility

5 / 5

About this paper

Methodology: ReSGA (Retrieval-Enhanced Self-Grouping Autoencoder). Problem types: Time Series Forecasting, Risk Management, Transfer Learning, Portfolio Optimization, Regression, Clustering.

arXiv:2606.04576 · Code · Paper rankings

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