HFuzzer: Testing Large Language Models for Package Hallucinations via Phrase-based Fuzzing
This addresses a critical security risk for software developers and organizations using LLMs in production, as package hallucinations can be exploited in software supply chain attacks, though it is incremental as it builds on fuzzing techniques for a specific domain.
The paper tackles the problem of package hallucinations in large language models (LLMs) used for code generation, where models recommend non-existent packages, by proposing HFUZZER, a phrase-based fuzzing framework that identifies 2.60x more unique hallucinated packages than a mutational fuzzing baseline and finds 46 unique hallucinated packages in GPT-4o.
Large Language Models (LLMs) are widely used for code generation, but they face critical security risks when applied to practical production due to package hallucinations, in which LLMs recommend non-existent packages. These hallucinations can be exploited in software supply chain attacks, where malicious attackers exploit them to register harmful packages. It is critical to test LLMs for package hallucinations to mitigate package hallucinations and defend against potential attacks. Although researchers have proposed testing frameworks for fact-conflicting hallucinations in natural language generation, there is a lack of research on package hallucinations. To fill this gap, we propose HFUZZER, a novel phrase-based fuzzing framework to test LLMs for package hallucinations. HFUZZER adopts fuzzing technology and guides the model to infer a wider range of reasonable information based on phrases, thereby generating enough and diverse coding tasks. Furthermore, HFUZZER extracts phrases from package information or coding tasks to ensure the relevance of phrases and code, thereby improving the relevance of generated tasks and code. We evaluate HFUZZER on multiple LLMs and find that it triggers package hallucinations across all selected models. Compared to the mutational fuzzing framework, HFUZZER identifies 2.60x more unique hallucinated packages and generates more diverse tasks. Additionally, when testing the model GPT-4o, HFUZZER finds 46 unique hallucinated packages. Further analysis reveals that for GPT-4o, LLMs exhibit package hallucinations not only during code generation but also when assisting with environment configuration.