BERT2Code: Can Pretrained Language Models be Leveraged for Code Search?

TL;DR

This paper explores leveraging pretrained language models for semantic code search by using embeddings and a simple neural network, highlighting the importance of embedding quality for search effectiveness.

Contribution

It demonstrates that a lightweight neural network can effectively utilize code and natural language embeddings for semantic code search, emphasizing the role of embedding quality.

Findings

Embedding quality is the main bottleneck for search performance.

A simple 2-layer neural network can learn relationships between embeddings.

Analysis suggests improving embedding models could enhance code search results.

Abstract

Millions of repetitive code snippets are submitted to code repositories every day. To search from these large codebases using simple natural language queries would allow programmers to ideate, prototype, and develop easier and faster. Although the existing methods have shown good performance in searching codes when the natural language description contains keywords from the code, they are still far behind in searching codes based on the semantic meaning of the natural language query and semantic structure of the code. In recent years, both natural language and programming language research communities have created techniques to embed them in vector spaces. In this work, we leverage the efficacy of these embedding models using a simple, lightweight 2-layer neural network in the task of semantic code search. We show that our model learns the inherent relationship between the embedding spaces and further probes into the scope of improvement by empirically analyzing the embedding methods. In this analysis, we show that the quality of the code embedding model is the bottleneck for our model's performance, and discuss future directions of study in this area.