Year: 2024
Type: conference-paper
Abstract: AI pair programmers, such as GitHub's Copilot, have shown great success in automatic code generation. However, such large language model-based code generation techniques face the risk of introducing security vulnerabilities to codebases. In this work, we explore the direction of fine-tuning large language models for generating more secure code. We use real-world vulnerability fixes as our fine-tuning dataset. We craft a code-generation scenario dataset (C/C++) for evaluating and comparing the pre-trained and fine-tuned models. Our experiments on GPT-J show that the fine-tuned GPT-J achieved 70.4% and 64.5% ratios of non-vulnerable code generation for C and C++, respectively, which has a 10% increase for C and a slight increase for C++ compared with the pre-trained large language model.
Authors: Junjie Li, Aseem Sangalay, Cheng Cheng, Yuan Tian, Jinqiu Yang
Language: English
Subfield: Information Systems
Field: Computer Science
Domain: Physical Sciences
Open Access status: bronze
OpenAlex ID: W4399567378