Disassembly as Weighted Interval Scheduling with Learned Weights

TL;DR

This paper introduces a generalized disassembly algorithm applicable to multiple architectures, framing conflict resolution as a weighted interval scheduling problem with learned weights, improving accuracy in binary analysis tasks.

Contribution

It presents a novel disassembly approach that models conflict resolution as weighted interval scheduling, applicable across various architectures, with a learned weighting scheme.

Findings

Effective disassembly across x86, x64, arm32, aarch64 architectures.

Improved disassembly accuracy through weighted interval scheduling.

Generalization of disassembly process for diverse architectures.

Abstract

Disassembly is the first step of a variety of binary analysis and transformation techniques, such as reverse engineering, or binary rewriting. Recent disassembly approaches consist of three phases: an exploration phase, that overapproximates the binary's code; an analysis phase, that assigns weights to candidate instructions or basic blocks; and a conflict resolution phase, that downselects the final set of instructions. We present a disassembly algorithm that generalizes this pattern for a wide range of architectures, namely x86, x64, arm32, and aarch64. Our algorithm presents a novel conflict resolution method that reduces disassembly to weighted interval scheduling.