Improving Retrieval-Augmented Deep Assertion Generation via Joint Training

TL;DR

This paper introduces AG-RAG, a retrieval-augmented assertion generation method that jointly trains retrieval and generation components using external codebases and advanced language models, significantly improving accuracy over prior approaches.

Contribution

The paper presents a novel joint training framework for assertion retrieval and generation, leveraging external codebases and CodeT5 to enhance assertion accuracy in unit testing.

Findings

AG-RAG outperforms six state-of-the-art approaches on multiple benchmarks.

It improves assertion accuracy by over 20% compared to the latest baseline.

AG-RAG generates more unique assertions than previous methods, with 3.45 to 9.20 times more unique assertions.

Abstract

Unit testing attempts to validate the correctness of basic units of the software system under test and has a crucial role in software development and testing. Very recent work proposes a retrieve-and-edit approach to generate unit test oracles, i.e., assertions. Despite being promising, it is still far from perfect due to some limitations, such as splitting assertion retrieval and generation into two separate components without benefiting each other. In this paper, we propose AG-RAG, a retrieval-augmented automated assertion generation approach that leverages external codebases and joint training to address various technical limitations of prior work. Inspired by the plastic surgery hypothesis, AG-RAG attempts to combine relevant unit tests and advanced pre-trained language models (PLMs) with retrieval-augmented fine-tuning. AG-RAG builds a dense retriever to search for relevant test-assert pairs (TAPs) with semantic matching and a retrieval-augmented generator to synthesize accurate assertions with the focal-test and retrieved TAPs as input. Besides, AG-RAG leverages a code-aware language model CodeT5 as the cornerstone to facilitate both assertion retrieval and generation tasks. Furthermore, the retriever is optimized in conjunction with the generator as a whole pipeline with a joint training strategy. This unified design fully adapts both components specifically for retrieving more useful TAPs, thereby generating accurate assertions. We extensively evaluate AG-RAG against six state-of-the-art AG approaches on two benchmarks and three metrics. Experimental results show that AG-RAG significantly outperforms previous AG approaches on all benchmarks and metrics, e.g., improving the most recent baseline EditAS by 20.82% and 26.98% in terms of accuracy. AG-RAG also correctly generates 1739 and 2866 unique assertions that all baselines fail to generate, 3.45X and 9.20X more than EditAS.