Retrieval-Based Neural Code Generation

Citations Over TimeTop 10% of 2018 papers

Abstract

In models to generate program source code from natural language, representing this code in a tree structure has been a common approach. However, existing methods often fail to generate complex code correctly due to a lack of ability to memorize large and complex structures. We introduce RECODE, a method based on subtree retrieval that makes it possible to explicitly reference existing code examples within a neural code generation model. First, we retrieve sentences that are similar to input sentences using a dynamicprogramming-based sentence similarity scoring method. Next, we extract n-grams of action sequences that build the associated abstract syntax tree. Finally, we increase the probability of actions that cause the retrieved n-gram action subtree to be in the predicted code. We show that our approach improves the performance on two code generation tasks by up to +2.6 BLEU. 1

Related Papers

• → Generalized vulnerability extrapolation using abstract syntax trees(2012)228 cited
• → A Deep Learning Model for Source Code Generation(2019)15 cited
• → Source Code Plagiarism Detection Based on Abstract Syntax Tree Fingerprintings(2022)6 cited
• → StructCoder: Structure-Aware Transformer for Code Generation(2022)7 cited