Generating Highly Structured Inputs: A Survey
Yabin Li, Yinfan Qin, Jianwen Tian, Taotao Gu, Yuanping Nie, Xiaohui Kuang · 2021
Fuzzing is currently one of the most popular vulnerability mining technology. The test case generation technique plays a key role in fuzzing, and the quality of the generated test cases directly affects the effect of fuzzing. Some programs process highly structured inputs which need to satisfy clear syntactic and semantic constraints, but traditional fuzzers (such as AFL) are difficult to generate such inputs. In this paper, we focus on the techniques of generating highly structured inputs. We first systematically introduce the mainstream techniques of generating highly structured inputs, then discuss the key issues facing the current technique and main technique trends, finally summarize the relevant open-source tools to promote future research.