Code Generation as a Dual Task of Code Summarization
This work addresses the need for better automated software development tools by integrating two related tasks, though it is incremental as it builds on existing neural network approaches.
The paper tackled the problem of improving code summarization and code generation by exploiting their intuitive correlation, proposing a dual training framework that uses regularization on probability and attention weights, and experimental results on GitHub datasets show performance improvements over baselines.
Code summarization (CS) and code generation (CG) are two crucial tasks in the field of automatic software development. Various neural network-based approaches are proposed to solve these two tasks separately. However, there exists a specific intuitive correlation between CS and CG, which have not been exploited in previous work. In this paper, we apply the relations between two tasks to improve the performance of both tasks. In other words, exploiting the duality between the two tasks, we propose a dual training framework to train the two tasks simultaneously. In this framework, we consider the dualities on probability and attention weights, and design corresponding regularization terms to constrain the duality. We evaluate our approach on two datasets collected from GitHub, and experimental results show that our dual framework can improve the performance of CS and CG tasks over baselines.