paper-with-me

홈 › Papers

Vulnerability Detection via Topological Analysis of Attention Maps

2024-10-04 · Pavel Snopov, Andrey Nikolaevich Golubinskiy

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.

📄 PDF Abstract BibTeX arXiv:2410.03470

Code (1)

Snopoff/Vulnerability-Detection-via-Topological-Analysis-of-Attention-Maps 공식 구현

Tasks

Topological Data AnalysisVulnerability Detection

Methods 이 논문이 사용한 방법론

Refunds@Expedia|||How do I get a full refund from Expedia? “How do I get a full refund from Expedia? How do I get a full refund from Expedia? – Call ☎️ +1-(888) 829 (0881) or +1-805-330-4056 or +1-805-330-4056 for Quick Help &…
Linear Layer A Linear Layer is a projection $\mathbf{XW + b}$.
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Multi-Head Attention 설명 없음
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Dense Connections Dense Connections, or Fully Connected Connections, are a type of layer in a deep neural network that use a linear operation where every input is connected to every output…
Adam 설명 없음
WordPiece 설명 없음

Similar Papers 제목 키워드 기반

Bridging Semantics & Structure for Software Vulnerability Detection using Hybrid Network Models

2025-10-11 · Jugal Gajjar, Kaustik Ranaware, Kamalasankari Subramaniakuppusamy arxiv

Software vulnerabilities remain a persistent risk, yet static and dynamic analyses often overlook structural dependencies that shape insecure behaviors. Viewing programs as heterogeneous graphs, we capture control- and d…

Vulnerability DetectionBinary Classification

Acceptability Judgements via Examining the Topology of Attention Maps

2022-05-19 · Daniil Cherniavskii, Eduard Tulchinskii, Vladislav Mikhailov, Irina Proskurina 외

The role of the attention mechanism in encoding linguistic knowledge has received special interest in NLP. However, the ability of the attention heads to judge the grammatical acceptability of a sentence has been underex…

CoLALinguistic AcceptabilitySentenceTopological Data Analysis

Uncertainty Estimation of Transformers' Predictions via Topological Analysis of the Attention Matrices

2023-08-22 · Elizaveta Kostenok, Daniil Cherniavskii, Alexey Zaytsev

Transformer-based language models have set new benchmarks across a wide range of NLP tasks, yet reliably estimating the uncertainty of their predictions remains a significant challenge. Existing uncertainty estimation (U…

text-classificationText ClassificationText DetectionTopological Data Analysis

Artificial Text Detection via Examining the Topology of Attention Maps

2021-09-10 · EMNLP 2021 11 · Laida Kushnareva, Daniil Cherniavskii, Vladislav Mikhailov, Ekaterina Artemova 외

The impressive capabilities of recent generative models to create texts that are challenging to distinguish from the human-written ones can be misused for generating fake news, product reviews, and even abusive content. …

Text DetectionTopological Data Analysis

HalluZig: Hallucination Detection using Zigzag Persistence

2026-01-04 · Shreyas N. Samaga, Gilberto Gonzalez Arroyo, Tamal K. Dey arxiv

The factual reliability of Large Language Models (LLMs) remains a critical barrier to their adoption in high-stakes domains due to their propensity to hallucinate. Current detection methods often rely on surface-level si…