.png){kind=icon}
返回
LMFuzz: Program repair fuzzing based on large language models
.png){kind=icon}
PRE
AI
DOI：10.1007/s10515-025-00568-8.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
Generating programs using large language models (LLMs) for fuzz testing has emerged as a significant testing methodology. While traditional fuzzers can produce correct programs, their effectiveness is limited by excessive constraints and restricted API combinations, resulting in insufficient coverage of the target system's code and impacting testing efficiency. Unlike traditional methods, large language model based fuzzers can generate more diverse code, effectively addressing key issues of conventional fuzzers. However, the lack of constraints on API combinations during the generation process often leads to reduced program validity. Therefore, a crucial challenge is to enhance the validity of generated code while maintaining its diversity. To address this issue, we propose a novel and universal fuzzer, LMFuzz. To ensure the fuzzer's generation capability, we utilize a large language model as the primary generator and model the operator selection problem within the fuzzing loop as a multi-armed bandit problem. We introduce the Thompson Sampling algorithm to enhance both the diversity and validity of program generation. To improve the validity of the generated code, we incorporate a program repair loop that iteratively corrects the generated programs, thereby reducing errors caused by the lack of API combination constraints. Experimental results demonstrate that LMFuzz significantly surpasses existing state-of-the-art large language model based fuzzers in terms of coverage and validity, and also exhibits notable advantages in generating diverse programs. Furthermore, LMFuzz has identified 24 bugs across five popular programming languages and their corresponding systems.
Keyword:
Fuzzing
Large language model
Code generation
Thompson sampling algorithm
Program Repair
Software security
期刊
A
IF:
3.1
论文数:
84
被引数:
0
机构
引用论文
.png){kind=icon}
Fuzzing JavaScript Interpreters with Coverage-Guided Reinforcement Learning for LLM-Based Mutation基于覆盖引导的强化学习的模糊JavaScript解释器，用于基于LLM的突变
PREAI
分享收藏CodeT5: Identifier-aware Unified Pre-trained Encoder-Decoder Models for Code Understanding and GenerationCodeT5: 用于代码理解和生成的标识符感知的统一预训练编码器-解码器模型
OAAI
OAAI分享收藏
查看更多内容