MPCODER: Multi-user Personalized Code Generator with Explicit and Implicit Style Representation Learning

TL;DR

MPCoder is a multi-user personalized code generator that learns explicit syntax and implicit semantic style features to generate user-specific code, using contrastive learning and a new evaluation metric.

Contribution

This paper introduces MPCoder, a novel approach that combines explicit and implicit style learning with contrastive adaptation for personalized code generation across multiple users.

Findings

Effective differentiation of user coding styles

Improved personalized code generation accuracy

Novel metric for style similarity evaluation

Abstract

Large Language Models (LLMs) have demonstrated great potential for assisting developers in their daily development. However, most research focuses on generating correct code, how to use LLMs to generate personalized code has seldom been investigated. To bridge this gap, we proposed MPCoder (Multi-user Personalized Code Generator) to generate personalized code for multiple users. To better learn coding style features, we utilize explicit coding style residual learning to capture the syntax code style standards and implicit style learning to capture the semantic code style conventions. We train a multi-user style adapter to better differentiate the implicit feature representations of different users through contrastive learning, ultimately enabling personalized code generation for multiple users. We further propose a novel evaluation metric for estimating similarities between codes of different coding styles. The experimental results show the effectiveness of our approach for this novel task.