Computer Science Seminar Series: Kexin Pei

Description

Kexin Pei, a fifth-year computer science PhD student at Columbia University, will give a talk titled "Scalable, Accurate, Robust Binary Analysis with Transfer Learning Trace Modeling" for the Department of Computer Science.

Please attend the event by using the Zoom link (meeting ID: 942 7282 5614).

Binary program analysis is a fundamental building block for a broad spectrum of security tasks, including vulnerability detection, reverse engineering, malware analysis, patching, security retrofitting, and forensics. Essentially, binary analysis encapsulates a diverse set of tasks that aim to understand and analyze how the binary program runs and its operational semantics. Unfortunately, existing approaches often tackle each analysis task independently and heavily employ ad-hoc heuristics as a shortcut for each task. These heuristics are often spurious and brittle, as they do not capture the real program semantics (behavior). While ML-based approaches have shown early promise, they too tend to learn spurious features and overfit specific tasks without understanding the underlying program semantics.

In this talk, Pei will describe two of his recent projects that learn program operational semantics for various binary analysis tasks. His key observation is that by designing pretraining tasks that can learn how binary programs execute, we can drastically boost the performance of binary analysis tasks. The pretraining tasks are fully self-supervised -- they do not need expensive labeling effort. Therefore, the pretrained models can use diverse binaries to generalize across different architectures, operating systems, compilers, and optimizations/obfuscations. Extensive experiments show that Pei's approach drastically improves the performance of tasks like matching semantically similar binary functions and binary type inference.