smq/directoryAn evolving quant resource archive
Paper / 2023SMQ / 145283752139619
Benchmarking Large Language Model Volatility
large language models / natural language processingResource overview

Benchmarking Large Language Model Volatility

Paper2023

by Boyang Yu

Tags
large language modelsnatural language processing

Continue on SelfMadeQuant

Understand this paper

Suggested foundations for text and language models in finance and volatility models and surfaces. Specialist prerequisites vary by paper.

Before you read

Go further

Explore a related reading path