JZUS - Journal of Zhejiang University SCIENCE

Frontiers of Information Technology & Electronic Engineering

Accepted manuscript available online (unedited version)

LLM-enhanced probabilistic modeling for effective static analysis alarms

Author(s): Xinlong PAN^1, ², Jianhua LI^1, ², Zhihong ZHOU^1, ², Gaolei LI^1, ², Xiuzhen CHEN^1, ², Jin MA^1, ², Jun WU^1, ², Quanhai ZHANG^1, ²
Affiliation(s): ¹Institute of Cyber Security and Technology, School of Computer Science, Shanghai Jiaotong University,Shanghai 200240, China; more
Corresponding email(s): mr.p332@sjtu.edu.cn
Key Words: Static analysis; Bayesian inference; large language model(LLM); Alarm ranking

Share this article to： More <<< Previous Paper \|Next Paper >>>

Xinlong PAN^1,², Jianhua LI^1,², Zhihong ZHOU^1,², Gaolei LI^1,², Xiuzhen CHEN^1,²,Jin MA^1,², Jun WU^1,², Quanhai ZHANG^1,². LLM-enhanced probabilistic modeling for effective static analysis alarms[J]. Frontiers of Information Technology & Electronic Engineering,in press.https://doi.org/10.1631/FITEE.2500038

@article{title="LLM-enhanced probabilistic modeling for effective static analysis alarms",
author="Xinlong PAN^1,², Jianhua LI^1,², Zhihong ZHOU^1,², Gaolei LI^1,², Xiuzhen CHEN^1,²,Jin MA^1,², Jun WU^1,², Quanhai ZHANG^1,²",
journal="Frontiers of Information Technology & Electronic Engineering",
year="in press",
publisher="Zhejiang University Press & Springer",
doi="https://doi.org/10.1631/FITEE.2500038"
}

%0 Journal Article
%T LLM-enhanced probabilistic modeling for effective static analysis alarms
%A Xinlong PAN^1
%A²
%A Jianhua LI^1
%A²
%A Zhihong ZHOU^1
%A²
%A Gaolei LI^1
%A²
%A Xiuzhen CHEN^1
%A²
%A Jin MA^1
%A²
%A Jun WU^1
%A²
%A Quanhai ZHANG^1
%A²
%J Frontiers of Information Technology & Electronic Engineering
%P
%@ 2095-9184
%D in press
%I Zhejiang University Press & Springer
doi="https://doi.org/10.1631/FITEE.2500038"

TY - JOUR
T1 - LLM-enhanced probabilistic modeling for effective static analysis alarms
A1 - Xinlong PAN^{1
A1 -}²
A1 - Jianhua LI^{1
A1 -}²
A1 - Zhihong ZHOU^{1
A1 -}²
A1 - Gaolei LI^{1
A1 -}²
A1 - Xiuzhen CHEN^{1
A1 -}²
A1 - Jin MA^{1
A1 -}²
A1 - Jun WU^{1
A1 -}²
A1 - Quanhai ZHANG^{1
A1 -}²
J0 - Frontiers of Information Technology & Electronic Engineering
SP -
EP -
%@ 2095-9184
Y1 - in press
PB - Zhejiang University Press & Springer
ER -
doi="https://doi.org/10.1631/FITEE.2500038"

Abstract
Chinese Summary
Academic Network
Reviewer Comment

Abstract: Static analysis presents significant challenges in alarm handling, where probabilistic models and alarm prioritization are essential methods for addressing these issues. These models prioritize alarms based on user feedback, thereby alleviating the burden on users to manually inspect alarms. However, they often encounter limitations related to efficiency and issues such as false generalization. While learning-based approaches have demonstrated promise, they typically incur high training costs and are constrained by the predefined structures of existing models. Moreover, the integration of large language models (LLMs) in static analysis has yet to reach its full potential, often resulting in lower accuracy rates in vulnerability identification.To tackle these challenges, we introduce BINLLM, a novel framework that harnesses the generalization capabilities of LLMs to enhance alarm probability models through rule learning. Our approach integrates LLM-derived abstract rules into the probabilistic model, utilizing alarm paths and critical statements from static analysis. This integration enhances the model’s reasoning capabilities, improving its effectiveness in prioritizing genuine bugs while mitigating false generalizations. We evaluated BINLLM on a suite of C programs and observed 40.1% and 9.4% reduction in the number of checks required for alarm verification compared to two state-of-the-art baselines, underscoring the potential of combining LLMs with static analysis to improve alarm management.

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference

Open peer comments: Debate/Discuss/Question/Opinion

<1>

- Go to

Darkslateblue:Affiliate; Royal Blue:Author; Turquoise:Article

Reference