D-2A

The 'D-2A' dataset/benchmark contains a collection of code snippets and associated warnings used to evaluate the effectiveness of Large Language Models (LLMs) in automating false positive mitigation in Static Application Security Testing (SAST) tools.

Papers using D-2A (4)

Utilizing Precise and Complete Code Context to Guide LLM in Automatic False Positive Mitigation2024 · 2 citesSAFE: Advancing Large Language Models in Leveraging Semantic and Syntactic Relationships for Software Vulnerability Detection2024D2A: A Dataset Built for AI-Based Vulnerability Detection Methods Using Differential Analysis2021 · 135 citesVulBERTa: Simplified Source Code Pre-Training for Vulnerability Detection2022 · 135 cites