This site is under active development. Some content may be AI-assisted or incomplete. If something looks off, it probably is.

Finance Language Model Evaluation (FLaME)

NLP
Finance
ACL
A comprehensive evaluation framework for studying language models against reasoning-reinforced LMs across 20+ core NLP tasks in finance.
Authors

Glenn Matlin

Mika Okamoto

Huzaifa Pardawala

Yang Yang

Sudheer Chava

Published

June 18, 2025

Publication Details

Venue: Findings of the Association for Computational Linguistics: ACL 2025

arXiv: 2506.15846

DOI: 10.18653/v1/2025.findings-acl.1164

Abstract

This is the first research paper to comprehensively study language models against reasoning-reinforced LMs, with an empirical study of 23 foundation LMs over 20 core NLP tasks in finance. FLaME provides a grounded evaluation framework that reveals significant differences in financial reasoning, with leading models achieving accuracy levels near 80%.

Back to top