A Neural Architecture for Generating Natural Language Descriptions from Source Code Changes
This work addresses the challenge of documenting code changes for developers, though it appears incremental as it builds on existing encoder-decoder architectures.
The authors tackled the problem of automatically generating natural language descriptions for source code changes by training an encoder-decoder model on code commits and user messages. Their approach achieved feasible and semantically sound descriptions in both in-project and cross-project settings, as validated on twelve open-source projects across four programming languages.
We propose a model to automatically describe changes introduced in the source code of a program using natural language. Our method receives as input a set of code commits, which contains both the modifications and message introduced by an user. These two modalities are used to train an encoder-decoder architecture. We evaluated our approach on twelve real world open source projects from four different programming languages. Quantitative and qualitative results showed that the proposed approach can generate feasible and semantically sound descriptions not only in standard in-project settings, but also in a cross-project setting.