When Fuzzing Meets LLMs: Challenges and Opportunities

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Jiang, Yu, Liang, Jie, Ma, Fuchen, Chen, Yuanliang, Zhou, Chijin, Shen, Yuheng, Wu, Zhiyong, Fu, Jingzhou, Wang, Mingzhe, Li, ShanShan, Zhang, Quan
Format: Preprint
Published: 2024
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866913329009655808
author Jiang, Yu
Liang, Jie
Ma, Fuchen
Chen, Yuanliang
Zhou, Chijin
Shen, Yuheng
Wu, Zhiyong
Fu, Jingzhou
Wang, Mingzhe
Li, ShanShan
Zhang, Quan
author_facet Jiang, Yu
Liang, Jie
Ma, Fuchen
Chen, Yuanliang
Zhou, Chijin
Shen, Yuheng
Wu, Zhiyong
Fu, Jingzhou
Wang, Mingzhe
Li, ShanShan
Zhang, Quan
contents 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.
format Preprint
id arxiv_https___arxiv_org_abs_2404_16297
institution arXiv
publishDate 2024
record_format arxiv
spellingShingle When Fuzzing Meets LLMs: Challenges and Opportunities
Jiang, Yu
Liang, Jie
Ma, Fuchen
Chen, Yuanliang
Zhou, Chijin
Shen, Yuheng
Wu, Zhiyong
Fu, Jingzhou
Wang, Mingzhe
Li, ShanShan
Zhang, Quan
Software Engineering
Artificial Intelligence
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.
title When Fuzzing Meets LLMs: Challenges and Opportunities
topic Software Engineering
Artificial Intelligence
url https://arxiv.org/abs/2404.16297