Design choices made by LLM-based test generators prevent them from finding bugs

TL;DR

This paper critically evaluates LLM-based test generators, revealing that their design choices often prevent bug detection and can validate faulty code, raising concerns about their effectiveness in software testing.

Contribution

It provides an empirical analysis showing how current LLM-based test tools fail to detect bugs and may validate faulty code due to their test oracle design.

Findings

LLM-generated tests often fail to detect bugs

Test oracles designed to pass can validate bugs

Design flaws can lead to rejecting bug-revealing tests

Abstract

There is an increasing amount of research and commercial tools for automated test case generation using Large Language Models (LLMs). This paper critically examines whether recent LLM-based test generation tools, such as Codium CoverAgent and CoverUp, can effectively find bugs or unintentionally validate faulty code. Considering bugs are only exposed by failing test cases, we explore the question: can these tools truly achieve the intended objectives of software testing when their test oracles are designed to pass? Using real human-written buggy code as input, we evaluate these tools, showing how LLM-generated tests can fail to detect bugs and, more alarmingly, how their design can worsen the situation by validating bugs in the generated test suite and rejecting bug-revealing tests. These findings raise important questions about the validity of the design behind LLM-based test generation tools and their impact on software quality and test suite reliability.