Adversarial Robustness for Code

2020/02/11 by Pavol Bielik, Martin Vechev, Bielik, Pavol +1 · 9 citations
Computer Science · Mathematics · #Advanced Malware Detection Techniques #Adversarial Robustness in Machine Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Machine Learning (stat.ML) #Programming Languages (cs.PL) #Security and Verification in Computing #Software Engineering (cs.SE) #cs.LG #cs.PL #cs.SE #stat.ML

paper · pdf · doi:10.48550/arxiv.2002.04694

Proceedings of the 37th International Conference on Machine Learning, Online, PMLR 119, 2020

openalex publication_date 2020/02/11 · arxiv created 2020/08/15 · arxiv updated 2020/08/18 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

Abstract

Machine learning and deep learning in particular has been recently used to successfully address many tasks in the domain of code such as finding and fixing bugs, code completion, decompilation, type inference and many others. However, the issue of adversarial robustness of models for code has gone largely unnoticed. In this work, we explore this issue by: (i) instantiating adversarial attacks for code (a domain with discrete and highly structured inputs), (ii) showing that, similar to other domains, neural models for code are vulnerable to adversarial attacks, and (iii) combining existing and novel techniques to improve robustness while preserving high accuracy.

Citations

• Adaptive Computation Time for Recurrent Neural Networks
• SelectiveNet: A Deep Neural Network with an Integrated Reject Option
• Inferring Javascript types using Graph Neural Networks
• A Convolutional Attention Network for Extreme Summarization of Source Code
• Detecting Functionality-Specific Vulnerabilities via Retrieving Individual Functionality-Equivalent APIs in Open-Source Repositories
• Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning
• LambdaNet: Probabilistic Type Inference using Graph Neural Networks
• Robustness to Programmable String Transformations via Augmented Abstract Training

Cited by

• Rethinking the Capability of Fine-Tuned Language Models for Automated Vulnerability Repair
• Readability-Robust Code Summarization via Meta Curriculum Learning
• NatGVD: Natural Adversarial Example Attack towards Graph-based Vulnerability Detection
• Evaluating Program Semantics Reasoning with Type Inference in System F
• "Digital Camouflage": The LLVM Challenge in LLM-Based Malware Detection
• Generating Adversarial Computer Programs using Optimized Obfuscations
• STRATA: Simple, Gradient-Free Attacks for Models of Code
• Understanding neural code intelligence through program simplification
• The Hitchhiker's Guide to Program Analysis, Part II: Deep Thoughts by LLMs

Related

• Towards Deep Learning Models Resistant to Adversarial Attacks
• Adversarial Robustness of Deep Learning: Theory, Algorithms, and Applications
• Adversarial Examples for Models of Code
• Recent Advances in Understanding Adversarial Robustness of Deep Neural Networks
• Holistic Adversarial Robustness of Deep Learning Models