Abstract

Large-scale pre-trained models such as CodeBERT, GraphCodeBERT have earned widespread attention from both academia and industry. Attributed to the superior ability in code representation, they have been further applied in multiple downstream tasks such as clone detection, code search and code translation. However, it is also observed that these state-of-the-art pre-trained models are susceptible to adversarial attacks. The performance of these pre-trained models drops significantly with simple perturbations such as renaming variable names. This weakness may be inherited by their downstream models and thereby amplified at an unprecedented scale. To this end, we propose an approach namely ContraBERT that aims to improve the robustness of pre-trained models via contrastive learning. Specifically, we design nine kinds of simple and complex data augmentation operators on the programming language (PL) and natural language (NL) data to construct different variants. Furthermore, we continue to train the existing pre-trained models by masked language modeling (MLM) and contrastive pre-training task on the original samples with their augmented variants to enhance the robustness of the model. The extensive experiments demonstrate that ContraBERT can effectively improve the robustness of the existing pre-trained models. Further study also confirms that these robustness-enhanced models provide improvements as compared to original models over four popular downstream tasks.

Contribution

1. We present a framework ContraBERT that enhances the robustness of existing pre-trained models in the code scenario by the pre-training tasks of masked language modeling and contrastive learning on original samples as well as the augmented variants.

2. We design nine kinds of simple or complex data augmentation operators on the programming language (PL) and natural language sequence (NL). Each operator confirms its effectiveness to improve the model's robustness.

3. The broad research on four downstream tasks demonstrates that the robustness-enhanced models provide improvements as compared to the original models.