LLM-Driven Adaptive Source-Sink Identification and False Positive Mitigation for Static Analysis
By: Shiyin Lin
Potential Business Impact:
Finds hidden computer bugs more accurately.
Static analysis is effective for discovering software vulnerabilities but notoriously suffers from incomplete source--sink specifications and excessive false positives (FPs). We present \textsc{AdaTaint}, an LLM-driven taint analysis framework that adaptively infers source/sink specifications and filters spurious alerts through neuro-symbolic reasoning. Unlike LLM-only detectors, \textsc{AdaTaint} grounds model suggestions in program facts and constraint validation, ensuring both adaptability and determinism. We evaluate \textsc{AdaTaint} on Juliet 1.3, SV-COMP-style C benchmarks, and three large real-world projects. Results show that \textsc{AdaTaint} reduces false positives by \textbf{43.7\%} on average and improves recall by \textbf{11.2\%} compared to state-of-the-art baselines (CodeQL, Joern, and LLM-only pipelines), while maintaining competitive runtime overhead. These findings demonstrate that combining LLM inference with symbolic validation offers a practical path toward more accurate and reliable static vulnerability analysis.
Similar Papers
Multi-Agent Taint Specification Extraction for Vulnerability Detection
Cryptography and Security
Finds hidden computer bugs using smart AI.
The Hitchhiker's Guide to Program Analysis, Part II: Deep Thoughts by LLMs
Software Engineering
Finds software bugs better by checking code carefully.
Learning to Triage Taint Flows Reported by Dynamic Program Analysis in Node.js Packages
Cryptography and Security
Helps find computer bugs faster and easier.