CodeFort: Robust Training for Code Generation Models

Code generation models are not robust to small perturbations, which often lead to incorrect generations and significantly degrade the performance of these models. Although improving the robustness of code generation models is crucial to enhancing user experience in real-world applications, existing...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2024-10
Main Authors: Zhang, Yuhao, Wang, Shiqi, Qian, Haifeng, Wang, Zijian, Shang, Mingyue, Liu, Linbo, Gouda, Sanjay Krishna, Ray, Baishakhi, Ramanathan, Murali Krishna, Ma, Xiaofei, Deoras, Anoop
Format: Article
Language:English
Subjects:
Data augmentation
Performance degradation
Perturbation
Robustness
User experience
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:Code generation models are not robust to small perturbations, which often lead to incorrect generations and significantly degrade the performance of these models. Although improving the robustness of code generation models is crucial to enhancing user experience in real-world applications, existing research efforts do not address this issue. To fill this gap, we propose CodeFort, a framework to improve the robustness of code generation models, generalizing a large variety of code perturbations to enrich the training data and enabling various robust training strategies, mixing data augmentation, batch augmentation, adversarial logits pairing, and contrastive learning, all carefully designed to support high-throughput training. Extensive evaluations show that we increase the average robust pass rates of baseline CodeGen models from 14.79 to 21.74. We notably decrease the robustness drop rate from 95.02% to 54.95% against code-syntax perturbations.
ISSN:2331-8422