Benchmarking Large Language Models on CFLUE -- A Chinese Financial Language Understanding Evaluation Dataset

TL;DR

This paper introduces CFLUE, a comprehensive Chinese financial language understanding benchmark for evaluating large language models across knowledge and application tasks, revealing current model limitations and progress.

Contribution

The paper presents CFLUE, a new benchmark with extensive datasets for Chinese financial NLP tasks, enabling systematic evaluation of LLMs' knowledge and application capabilities.

Findings

GPT-4 surpasses 60% accuracy in knowledge assessment

GPT-4 and GPT-4-turbo outperform lightweight LLMs in application tasks

Current LLMs still have significant room for improvement in financial NLP

Abstract

In light of recent breakthroughs in large language models (LLMs) that have revolutionized natural language processing (NLP), there is an urgent need for new benchmarks to keep pace with the fast development of LLMs. In this paper, we propose CFLUE, the Chinese Financial Language Understanding Evaluation benchmark, designed to assess the capability of LLMs across various dimensions. Specifically, CFLUE provides datasets tailored for both knowledge assessment and application assessment. In knowledge assessment, it consists of 38K+ multiple-choice questions with associated solution explanations. These questions serve dual purposes: answer prediction and question reasoning. In application assessment, CFLUE features 16K+ test instances across distinct groups of NLP tasks such as text classification, machine translation, relation extraction, reading comprehension, and text generation. Upon CFLUE, we conduct a thorough evaluation of representative LLMs. The results reveal that only GPT-4 and GPT-4-turbo achieve an accuracy exceeding 60\% in answer prediction for knowledge assessment, suggesting that there is still substantial room for improvement in current LLMs. In application assessment, although GPT-4 and GPT-4-turbo are the top two performers, their considerable advantage over lightweight LLMs is noticeably diminished. The datasets and scripts associated with CFLUE are openly accessible at https://github.com/aliyun/cflue.