AstBERT: Enabling Language Model for Financial Code Understanding with Abstract Syntax Trees

TL;DR

AstBERT is a specialized pre-trained language model that leverages abstract syntax trees to improve understanding of financial source code, enhancing performance in code-related tasks.

Contribution

The paper introduces AstBERT, a novel model integrating AST information into pre-trained language models for better financial code understanding.

Findings

AstBERT outperforms baseline models on code question answering.

AstBERT achieves higher accuracy in code clone detection.

AstBERT shows promising results in code refinement tasks.

Abstract

Using the pre-trained language models to understand source codes has attracted increasing attention from financial institutions owing to the great potential to uncover financial risks. However, there are several challenges in applying these language models to solve programming language-related problems directly. For instance, the shift of domain knowledge between natural language (NL) and programming language (PL) requires understanding the semantic and syntactic information from the data from different perspectives. To this end, we propose the AstBERT model, a pre-trained PL model aiming to better understand the financial codes using the abstract syntax tree (AST). Specifically, we collect a sheer number of source codes (both Java and Python) from the Alipay code repository and incorporate both syntactic and semantic code knowledge into our model through the help of code parsers, in which AST information of the source codes can be interpreted and integrated. We evaluate the performance of the proposed model on three tasks, including code question answering, code clone detection and code refinement. Experiment results show that our AstBERT achieves promising performance on three different downstream tasks.