When Fuzzing Meets LLMs: Challenges and Opportunities
This work addresses challenges in applying LLMs to fuzzing, which is incremental as it builds on existing advancements in LLM-assisted fuzzing.
The paper identifies five major challenges in using Large Language Models (LLMs) for fuzzing, a bug detection technique, and proposes actionable recommendations that show effectiveness in preliminary evaluations on DBMS fuzzing.
Fuzzing, a widely-used technique for bug detection, has seen advancements through Large Language Models (LLMs). Despite their potential, LLMs face specific challenges in fuzzing. In this paper, we identified five major challenges of LLM-assisted fuzzing. To support our findings, we revisited the most recent papers from top-tier conferences, confirming that these challenges are widespread. As a remedy, we propose some actionable recommendations to help improve applying LLM in Fuzzing and conduct preliminary evaluations on DBMS fuzzing. The results demonstrate that our recommendations effectively address the identified challenges.