一句话：提出基于专家特征知识的浅层神经网络组合框架，提升代码作者归属对抗扰动鲁棒性。
代码作者归属方法在攻击者添加微小扰动时鲁棒性较差，现有方法难以有效应对。本文将弱鲁棒性归因于数据集偏差，认为可通过调整特征学习策略来缓解。作者提出一个鲁棒的代码作者归属特征组合框架，该框架仅由简单浅层神经网络构成，并在特征提取中引入专家知识以增强可控性。实验表明，相比主流代码作者归属方法，该框架显著提升了鲁棒性：定向攻击成功率平均下降23.4%（从37.8%降至14.3%），非定向攻击成功率平均下降25.9%（从46.7%降至20.8%），同时在准确率上可达到与主流方法相当的结果。
Code authorship attribution has been an interesting research problem for decades. Recent studies have revealed that existing methods for code authorship attribution suffer from weak robustness. Under the influence of small perturbations added by the attacker, the accuracy of the method will be greatly reduced. As of now, there is no code authorship attribution method capable of effectively handling such attacks. In this paper, we attribute the weak robustness of code authorship attribution methods to dataset bias and argue that this bias can be mitigated through adjustments to the feature learning strategy. We first propose a robust code authorship attribution feature combination framework, which is composed of only simple shallow neural network structures, and introduces controllability for the framework in the feature extraction by incorporating expert knowledge. Experiments show that the framework has significantly improved robustness over mainstream code authorship attribution methods, with an average drop of 23.4% (from 37.8% to 14.3%) in the success rate of targeted attacks and 25.9% (from 46.7% to 20.8%) in the success rate of untargeted attacks. At the same time, it can also achieve results comparable to mainstream code authorship attribution methods in terms of accuracy.
DOI 原文 ·
@inproceedings{paperbot8546,
title = {Enhancing Robustness of Code Authorship Attribution through Expert Feature Knowledge},
author = {Xiaowei Guo and Cai Fu and Juan Chen and Hongle Liu and Lansheng Han and Wenjin Li},
booktitle = {ISSTA},
year = {2024},
doi = {10.1145/3650212.3652121}
}