Core Concepts
FineWAVE proposes a novel approach for fine-grained bug-sensitive warning verification, significantly improving the effectiveness of ASATs.
Abstract
The content discusses the challenges in verifying warnings generated by Automated Static Analysis Tools (ASATs) and introduces FineWAVE, a new approach for bug-sensitive warning verification. It highlights the limitations of existing methods, the dataset creation process, model architecture, and experimental results comparing FineWAVE with baseline models.
Structure:
Introduction to ASATs and the need for bug detection.
Challenges with false positives in ASAT warnings.
Introduction of FineWAVE approach for fine-grained warning verification.
Dataset creation process and model architecture.
Experimental results comparing FineWAVE with baseline models.
Evaluation of dataset quality through manual assessment.
Stats
The experimental results demonstrate an F1-score of 97.79% for reducing false alarms and 67.06% for confirming actual warnings.
Quotes
"We proposed a fine-grained warning verification approach that is sensitive to bugs for improving the results of ASATs."
"FineWAVE helps filter out 92% of the warnings in real-world projects."