Nimbus: Toward Speed Up Function Signature Recovery via Input Resizing and Multi-Task Learning

TL;DR

Nimbus is an efficient method for function signature recovery that significantly reduces resource consumption while maintaining or slightly improving accuracy through input resizing and multi-task learning.

Contribution

The paper introduces Nimbus, a novel approach combining input resizing and multi-task learning to enhance efficiency in function signature recovery without sacrificing accuracy.

Findings

Uses only about one-eighth of the processing time of state-of-the-art methods.

Achieves approximately 1% higher prediction accuracy.

Effectively leverages mutual information for improved performance.

Abstract

Function signature recovery is important for many binary analysis tasks such as control-flow integrity enforcement, clone detection, and bug finding. Existing works try to substitute learning-based methods with rule-based methods to reduce human effort.They made considerable efforts to enhance the system's performance, which also bring the side effect of higher resource consumption. However, recovering the function signature is more about providing information for subsequent tasks, and both efficiency and performance are significant. In this paper, we first propose a method called Nimbus for efficient function signature recovery that furthest reduces the whole-process resource consumption without performance loss. Thanks to information bias and task relation (i.e., the relation between parameter count and parameter type recovery), we utilize selective inputs and introduce multi-task learning (MTL) structure for function signature recovery to reduce computational resource consumption, and fully leverage mutual information. Our experimental results show that, with only about the one-eighth processing time of the state-of-the-art method, we even achieve about 1% more prediction accuracy over all function signature recovery tasks.