Learning to Reduce False Positives in Analytic Bug Detectors

TL;DR

This paper introduces a Transformer-based learning method to reduce false positives in static analysis bug warnings, significantly improving precision and demonstrating generalizability across multiple bug types.

Contribution

The paper presents a novel Transformer-based approach to accurately identify false positives in static analysis tools, enhancing bug detection precision.

Findings

Improved static analysis precision by 17.5%

Effective across null dereference and resource leak bugs

Demonstrated generalizability of the approach

Abstract

Due to increasingly complex software design and rapid iterative development, code defects and security vulnerabilities are prevalent in modern software. In response, programmers rely on static analysis tools to regularly scan their codebases and find potential bugs. In order to maximize coverage, however, these tools generally tend to report a significant number of false positives, requiring developers to manually verify each warning. To address this problem, we propose a Transformer-based learning approach to identify false positive bug warnings. We demonstrate that our models can improve the precision of static analysis by 17.5%. In addition, we validated the generalizability of this approach across two major bug types: null dereference and resource leak.