Enregistré dans:
Détails bibliographiques
Auteurs principaux: Shan, Zhengyang, Qian, Xu, Xin, Jiayun, Xu, Minghui, Zhang, Yue, Yang, Zhen, Wu, Hao, Cheng, Xiuzhen
Format: Preprint
Publié: 2026
Sujets:
Accès en ligne:https://arxiv.org/abs/2604.19031
Tags: Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
_version_ 1866908982724001792
author Shan, Zhengyang
Qian, Xu
Xin, Jiayun
Xu, Minghui
Zhang, Yue
Yang, Zhen
Wu, Hao
Cheng, Xiuzhen
author_facet Shan, Zhengyang
Qian, Xu
Xin, Jiayun
Xu, Minghui
Zhang, Yue
Yang, Zhen
Wu, Hao
Cheng, Xiuzhen
contents Software vulnerabilities are a primary threat to modern infrastructure. While static analysis and Graph Neural Networks have long served as the foundation for vulnerability detection, the emergence of Large Language Models (LLMs) has introduced a transformative paradigm driven by superior semantic reasoning and cross-environment generalization. However, in the context of LLM-based vulnerability detection, we identify a fundamental bottleneck in these models termed \textbf{Signal Submersion}: a state where features related to vulnerability are activated internally but numerically overwhelmed by dominant functional semantics. To address this, we propose \textbf{SAGE} (\textbf{S}ignal-\textbf{A}mplified \textbf{G}uided \textbf{E}mbeddings), a framework that shifts from passive signal submersion to active signal recovery. SAGE integrates task-conditional Sparse Autoencoders (SAEs) to isolate and amplify these faint vulnerability signals. Extensive evaluations on BigVul, PrimeVul, and PreciseBugs demonstrate that SAGE achieves state-of-the-art performance. Notably, SAGE mitigates Signal Submersion by increasing the internal Signal-to-Noise Ratio (SNR) by 12.7$\times$ via sparse manifold projection. This mechanistic intervention enables a 7B model to achieve up to 318\% Matthews Correlation Coefficient (MCC) gains on unseen distributions and a 319\% gain on classic datasets. By maintaining robust performance across 13 programming languages and outperforming 34B baselines, SAGE establishes a more efficient and scalable path to software security than simple parameter scaling.
format Preprint
id arxiv_https___arxiv_org_abs_2604_19031
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle SAGE: Signal-Amplified Guided Embeddings for LLM-based Vulnerability Detection
Shan, Zhengyang
Qian, Xu
Xin, Jiayun
Xu, Minghui
Zhang, Yue
Yang, Zhen
Wu, Hao
Cheng, Xiuzhen
Cryptography and Security
Software vulnerabilities are a primary threat to modern infrastructure. While static analysis and Graph Neural Networks have long served as the foundation for vulnerability detection, the emergence of Large Language Models (LLMs) has introduced a transformative paradigm driven by superior semantic reasoning and cross-environment generalization. However, in the context of LLM-based vulnerability detection, we identify a fundamental bottleneck in these models termed \textbf{Signal Submersion}: a state where features related to vulnerability are activated internally but numerically overwhelmed by dominant functional semantics. To address this, we propose \textbf{SAGE} (\textbf{S}ignal-\textbf{A}mplified \textbf{G}uided \textbf{E}mbeddings), a framework that shifts from passive signal submersion to active signal recovery. SAGE integrates task-conditional Sparse Autoencoders (SAEs) to isolate and amplify these faint vulnerability signals. Extensive evaluations on BigVul, PrimeVul, and PreciseBugs demonstrate that SAGE achieves state-of-the-art performance. Notably, SAGE mitigates Signal Submersion by increasing the internal Signal-to-Noise Ratio (SNR) by 12.7$\times$ via sparse manifold projection. This mechanistic intervention enables a 7B model to achieve up to 318\% Matthews Correlation Coefficient (MCC) gains on unseen distributions and a 319\% gain on classic datasets. By maintaining robust performance across 13 programming languages and outperforming 34B baselines, SAGE establishes a more efficient and scalable path to software security than simple parameter scaling.
title SAGE: Signal-Amplified Guided Embeddings for LLM-based Vulnerability Detection
topic Cryptography and Security
url https://arxiv.org/abs/2604.19031