Vulnerability Detection via Topological Analysis of Attention Maps
Fuente:
arXiv
Guardado en:
| Autores principales: | , |
|---|---|
| Formato: | Preprint |
| Publicado: |
2024
|
| Materias: | |
| Acceso en línea: | |
| Etiquetas: |
Agregar Etiqueta
Sin Etiquetas, Sea el primero en etiquetar este registro!
|
| _version_ | 1866914964723204096 |
|---|---|
| author | Snopov, Pavel Golubinskiy, Andrey Nikolaevich |
| author_facet | Snopov, Pavel Golubinskiy, Andrey Nikolaevich |
| contents | Recently, deep learning (DL) approaches to vulnerability detection have gained significant traction. These methods demonstrate promising results, often surpassing traditional static code analysis tools in effectiveness.
In this study, we explore a novel approach to vulnerability detection utilizing the tools from topological data analysis (TDA) on the attention matrices of the BERT model. Our findings reveal that traditional machine learning (ML) techniques, when trained on the topological features extracted from these attention matrices, can perform competitively with pre-trained language models (LLMs) such as CodeBERTa. This suggests that TDA tools, including persistent homology, are capable of effectively capturing semantic information critical for identifying vulnerabilities. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2410_03470 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Vulnerability Detection via Topological Analysis of Attention Maps Snopov, Pavel Golubinskiy, Andrey Nikolaevich Machine Learning Artificial Intelligence Algebraic Topology Recently, deep learning (DL) approaches to vulnerability detection have gained significant traction. These methods demonstrate promising results, often surpassing traditional static code analysis tools in effectiveness. In this study, we explore a novel approach to vulnerability detection utilizing the tools from topological data analysis (TDA) on the attention matrices of the BERT model. Our findings reveal that traditional machine learning (ML) techniques, when trained on the topological features extracted from these attention matrices, can perform competitively with pre-trained language models (LLMs) such as CodeBERTa. This suggests that TDA tools, including persistent homology, are capable of effectively capturing semantic information critical for identifying vulnerabilities. |
| title | Vulnerability Detection via Topological Analysis of Attention Maps |
| topic | Machine Learning Artificial Intelligence Algebraic Topology |
| url | https://arxiv.org/abs/2410.03470 |