Abstract:The rapid inflation of software features brings inefficiency and vulnerabilities into programs, resulting in an increased attack surface with a higher possibility of exploitation. In this paper, we propose a novel framework for automated software mass customization (AMASS), which automatically identifies program features from binaries, tailors and eliminates the features to create customized program binaries in accordance with user needs, in a fully unsupervised fashion. It enables us to modularize program features and efficiently create customized program binaries at large scale. Evaluation using real-world executables including OpenSSL and LibreOffice demonstrates that AMASS can create a wide range of customized binaries for diverse feature requirements, with an average 92.76% accuracy for feature/function identification and up to 67% reduction of program attack surface.

AMASS: Automated Software Mass Customization Via Feature Identification and Tailoring.