mlsanityData to 7 Oct 2026

Towards Making Deep Learning-based Vulnerability Detectors Robust

arXiv 2108.00669 ↗First posted 2 Aug 2021cs.CR cs.LGcited by 4

Not on the map. Why?

Abstract (arXiv)

Automatically detecting software vulnerabilities in source code is an important problem that has attracted much attention. In particular, deep learning-based vulnerability detectors, or DL-based detectors, are attractive because they do not need human experts to define features or patterns of vulnerabilities. However, such detectors' robustness is unclear. In this paper, we initiate the study in this aspect by demonstrating that DL-based detectors are not robust against simple code transformations, dubbed attacks in this paper, as these transformations may be leveraged for malicious purposes. As a first step towards making DL-based detectors robust against such attacks, we propose an innovative framework, dubbed ZigZag, which is centered at (i) decoupling feature learning and classifier learning and (ii) using a ZigZag-style strategy to iteratively refine them until they converge to robust features and robust classifiers. Experimental results show that the ZigZag framework can substantially improve the robustness of DL-based detectors.

Key facts. Cited by 4 papers in the map (data through 7 Oct 2026). Builds on 20 of its 73 references that are in the map. First posted on arXiv 2 Aug 2021.

Builds on

20 of 73 references.

References · 20

  1. 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
  2. 2019Adversarial Training for Free!cited by 638
  3. 2018TextBugger: Generating Adversarial Text Against Real-world Applicationscited by 376
  4. 2017Adversarial Examples: Attacks and Defenses for Deep Learningcited by 350
  5. 2019Devign: Effective Vulnerability Identification by Learning Comprehensive Program Semantics via Graph Neural Networkscited by 266
  6. 2018code2vec: Learning Distributed Representations of Codecited by 249
  7. 2019Imperceptible, Robust, and Targeted Adversarial Examples for Automatic Speech Recognitioncited by 123
  8. 2018Automated Vulnerability Detection in Source Code Using Deep Representation Learningcited by 114
  9. 2018SySeVR: A Framework for Using Deep Learning to Detect Software Vulnerabilitiescited by 91
  10. 2018The Art, Science, and Engineering of Fuzzing: A Surveycited by 61
  11. 2019Adversarial Examples for Models of Codecited by 61
  12. 2020Adversarial Robustness for Codecited by 36
8 more
  1. 2019Misleading Authorship Attribution of Source Code using Adversarial Learningcited by 33
  2. 2020Adversarial Deep Ensemble: Evasion Attacks and Defenses for Malware Detectioncited by 33
  3. 2019On Training Robust PDF Malware Classifierscited by 19
  4. 2018Learning to Repair Software Vulnerabilities with Generative Adversarial Networkscited by 11
  5. 2018Defense Against the Dark Arts: An overview of adversarial example security research and future research directionscited by 11
  6. 2019COSET: A Benchmark for Evaluating Neural Program Embeddingscited by 10
  7. 2020Robustness to Programmable String Transformations via Augmented Abstract Trainingcited by 8
  8. 2015Towards Vulnerability Discovery Using Staged Program Analysiscited by 1

Built on by

Cited by year

Citing papers per year: 2023: 2; 2024: 2.

4 papers.

  1. 2023Uncovering the Limits of Machine Learning for Automatic Vulnerability Detectioncited by 30
  2. 2023An Unbiased Transformer Source Code Learning with Semantic Vulnerability Graphcited by 5
  3. 2024Enhancing Source Code Security with LLMs: Demystifying The Challenges and Generating Reliable Repairscited by 4
  4. 2024Unintentional Security Flaws in Code: Automated Defense via Root Cause Analysiscited by 0