Saved in:
Bibliographic Details
Main Authors: Dhulipala, Hridya, Rong, Xiaokai, Nguyen, Tien N.
Format: Preprint
Published: 2025
Subjects:
Online Access:https://arxiv.org/abs/2512.21431
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866908731717976064
author Dhulipala, Hridya
Rong, Xiaokai
Nguyen, Tien N.
author_facet Dhulipala, Hridya
Rong, Xiaokai
Nguyen, Tien N.
contents In several software development scenarios, it is desirable to detect runtime errors and exceptions in code snippets without actual execution. A typical example is to detect runtime exceptions in online code snippets before integrating them into a codebase. In this paper, we propose Cerberus, a novel predictive, execution-free coverage-guided testing framework. Cerberus uses LLMs to generate the inputs that trigger runtime errors and to perform code coverage prediction and error detection without code execution. With a two-phase feedback loop, Cerberus first aims to both increasing code coverage and detecting runtime errors, then shifts to focus only detecting runtime errors when the coverage reaches 100% or its maximum, enabling it to perform better than prompting the LLMs for both purposes. Our empirical evaluation demonstrates that Cerberus performs better than conventional and learning-based testing frameworks for (in)complete code snippets by generating high-coverage test cases more efficiently, leading to the discovery of more runtime errors.
format Preprint
id arxiv_https___arxiv_org_abs_2512_21431
institution arXiv
publishDate 2025
record_format arxiv
spellingShingle Cerberus: Multi-Agent Reasoning and Coverage-Guided Exploration for Static Detection of Runtime Errors
Dhulipala, Hridya
Rong, Xiaokai
Nguyen, Tien N.
Software Engineering
Machine Learning
In several software development scenarios, it is desirable to detect runtime errors and exceptions in code snippets without actual execution. A typical example is to detect runtime exceptions in online code snippets before integrating them into a codebase. In this paper, we propose Cerberus, a novel predictive, execution-free coverage-guided testing framework. Cerberus uses LLMs to generate the inputs that trigger runtime errors and to perform code coverage prediction and error detection without code execution. With a two-phase feedback loop, Cerberus first aims to both increasing code coverage and detecting runtime errors, then shifts to focus only detecting runtime errors when the coverage reaches 100% or its maximum, enabling it to perform better than prompting the LLMs for both purposes. Our empirical evaluation demonstrates that Cerberus performs better than conventional and learning-based testing frameworks for (in)complete code snippets by generating high-coverage test cases more efficiently, leading to the discovery of more runtime errors.
title Cerberus: Multi-Agent Reasoning and Coverage-Guided Exploration for Static Detection of Runtime Errors
topic Software Engineering
Machine Learning
url https://arxiv.org/abs/2512.21431