Deep learning for coverage-guided fuzzing: How far are we?

• Main Authors: LI, Siqi, XIE, Xiaofei, LIN, Yun, LI, Yuekang, FENG, Ruitao, LI, Xiaohong, GE, Weimin, DONG, Jin Song
• Format: text
• Language: English
• Published: Institutional Knowledge at Singapore Management University 2022
• Subjects: Deep Learning; Testing; Fuzzing; Mutation; Coverage; Artificial Intelligence and Robotics
• Online Access: https://ink.library.smu.edu.sg/sis_research/7494; https://ink.library.smu.edu.sg/context/sis_research/article/8497/viewcontent/tdsc22_fuzzing.pdf
• Institution: Singapore Management University

Fuzzing is a widely-used software vulnerability discovery technology, many of which are optimized using coverage-feedback. Recently, some techniques propose to train deep learning (DL) models to predict the branch coverage of an arbitrary input owing to its always-available gradients etc. as a guide...