Semantic-Preserving Adversarial Code Comprehension
Not on the map. Why?
Abstract (arXiv)
Based on the tremendous success of pre-trained language models (PrLMs) for source code comprehension tasks, current literature studies either ways to further improve the performance (generalization) of PrLMs, or their robustness against adversarial attacks. However, they have to compromise on the trade-off between the two aspects and none of them consider improving both sides in an effective and practical way. To fill this gap, we propose Semantic-Preserving Adversarial Code Embeddings (SPACE) to find the worst-case semantic-preserving attacks while forcing the model to predict the correct labels under these worst cases. Experiments and analysis demonstrate that SPACE can stay robust against state-of-the-art attacks while boosting the performance of PrLMs for code.
Key facts. Cited by 2 papers in the map (data through 7 Oct 2026). Builds on 29 of its 36 references that are in the map. First posted on arXiv 12 Sep 2022.
Builds on
29 of 36 references.
References · 29
- 2017Attention Is All You Needcited by 59,932
- 2018BERT: Pre-training of Deep Bidirectional Transformers for Language Understandingcited by 40,204
- 2019RoBERTa: A Robustly Optimized BERT Pretraining Approachcited by 13,266
- 2014Explaining and Harnessing Adversarial Examplescited by 7,287
- 2017Towards Deep Learning Models Resistant to Adversarial Attackscited by 6,326
- 2015Neural Machine Translation of Rare Words with Subword Unitscited by 2,898
- 2019XLNet: Generalized Autoregressive Pretraining for Language Understandingcited by 2,801
- 2020CodeBERT: A Pre-Trained Model for Programming and Natural Languagescited by 1,286
- 2021CodeXGLUE: A Machine Learning Benchmark Dataset for Code Understanding and Generationcited by 732
- 2019CodeSearchNet Challenge: Evaluating the State of Semantic Code Searchcited by 638
- 2020GraphCodeBERT: Pre-training Code Representations with Data Flowcited by 600
- 2016Adversarial Training Methods for Semi-Supervised Text Classificationcited by 280
17 more
- 2019Devign: Effective Vulnerability Identification by Learning Comprehensive Program Semantics via Graph Neural Networkscited by 266
- 2019FreeLB: Enhanced Adversarial Training for Natural Language Understandingcited by 261
- 2019Robust Neural Machine Translation with Doubly Adversarial Inputscited by 105
- 2020Contrastive Code Representation Learningcited by 76
- 2022Natural Attack for Pre-trained Models of Codecited by 72
- 2018Interpretable Adversarial Perturbation in Input Embedding Space for Textcited by 67
- 2021SynCoBERT: Syntax-Guided Multi-Modal Contrastive Pre-Training for Code Representationcited by 63
- 2019Adversarial Examples for Models of Codecited by 61
- 2021CoTexT: Multi-task Learning with Code-Text Transformercited by 61
- 2020Self-Supervised Contrastive Learning for Code Retrieval and Summarization via Semantic-Preserving Transformationscited by 49
- 2020Semantic Robustness of Models of Source Codecited by 43
- 2020Adversarial Robustness for Codecited by 36
- 2019Learning and Evaluating Contextual Embedding of Source Codecited by 33
- 2020Exploring Software Naturalness through Neural Language Modelscited by 32
- 2021CodeQA: A Question Answering Dataset for Source Code Comprehensioncited by 30
- 2020Code Summarization with Structure-induced Transformercited by 29
- 2021Generating Adversarial Computer Programs using Optimized Obfuscationscited by 20
Built on by
Citing papers per year: 2023: 1; 2024: 0; 2025: 1.
2 papers.