Two Sides of the Same Coin: Exploiting the Impact of Identifiers in Neural Code Comprehension

TL;DR

This paper introduces CREAM, a novel framework that models the dual impact of identifiers in neural code comprehension using causal and counterfactual reasoning, significantly improving robustness and accuracy.

Contribution

The paper proposes a causal, counterfactual reasoning-based framework called CREAM to better exploit identifiers' dual effects in neural code comprehension models.

Findings

CREAM outperforms baselines in robustness (+37.9% F1 on function naming)

CREAM improves accuracy on original datasets (+0.5% F1)

Effective in tasks: function naming, defect detection, code classification

Abstract

Previous studies have demonstrated that neural code comprehension models are vulnerable to identifier naming. By renaming as few as one identifier in the source code, the models would output completely irrelevant results, indicating that identifiers can be misleading for model prediction. However, identifiers are not completely detrimental to code comprehension, since the semantics of identifier names can be related to the program semantics. Well exploiting the two opposite impacts of identifiers is essential for enhancing the robustness and accuracy of neural code comprehension, and still remains under-explored. In this work, we propose to model the impact of identifiers from a novel causal perspective, and propose a counterfactual reasoning-based framework named CREAM. CREAM explicitly captures the misleading information of identifiers through multi-task learning in the training stage, and reduces the misleading impact by counterfactual inference in the inference stage. We evaluate CREAM on three popular neural code comprehension tasks, including function naming, defect detection and code classification. Experiment results show that CREAM not only significantly outperforms baselines in terms of robustness (e.g., +37.9% on the function naming task at F1 score), but also achieve improved results on the original datasets (e.g., +0.5% on the function naming task at F1 score).