MergeBERT: Program Merge Conflict Resolution via Neural Transformers

Alexey Svyatkovskiy, Todd Mytkowicz, Negar Ghorbani, Sarah Fakhoury, Elizabeth A Dinella, Christian Bird, Neel Sundaresan, Shuvendu Lahiri

Keywords: Software evolution, program merge, ml4code

Abstract:

Collaborative software development is an integral part of the modern software development life cycle, essential to the success of large-scale software projects. When multiple developers make concurrent changes around the same lines of code, a merge conflict may occur.
Such conflicts stall pull requests and continuous integration pipelines for hours to several days, seriously hurting developer productivity.

In this paper, we introduce MergeBERT, a novel neural program merge framework based on the token-level three-way differencing and a transformer encoder model. Exploiting restricted nature of merge conflict resolutions, we reformulate the task of generating the resolution sequence as a classification task over a set of primitive merge patterns extracted from real-world merge commit data.

Our model achieves 63--68\% accuracy of merge resolution synthesis, yielding nearly a 3$\times$ performance improvement over existing structured, and 2$\times$ improvement over neural program merge tools. Finally, we demonstrate that MergeBERT is sufficiently flexible to work with source code files in Java, JavaScript, TypeScript, and C\# programming languages, and can generalize zero-shot to unseen languages.

One-sentence Summary: Program merge conflict resolution model based on BERT finetuned for sequence classification task.

Community Implementations: [ 2 code implementations](https://www.catalyzex.com/paper/arxiv:2109.00084/code)