# VulnScout-C: A Lightweight Transformer for C Code Vulnerability Detection

Source: [VulnScout-C: A Lightweight Transformer for C Code Vulnerability Detection](<https://www.sota2.com/research/paper/69cb55760636e5bf4afca347>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-04-01T05:00:12.613Z

Authors: Aymen Lassoued, Nacef Mbarek, Bechir Dardouri, Bassem Ouni, Qing Li, Fakhri Karray

Published: 2026-03-30T11:33:32.000Z

## Abstract

Vulnerability detection in C programs is a critical challenge in software security. Although large language models (LLMs) achieve strong detection performance, their multi-billion-parameter scale makes them impractical for integration into development workflows requiring low latency and continuous analysis. We introduce VULNSCOUT-C, a compact transformer architecture with 693M total parameters (353M active during inference), derived from the Qwen model family and optimized for C code vulnerability detection. Alongside the model, we present VULNSCOUT, a new 33,565-sample curated dataset generated through a controlled multi-agent pipeline with formal verification, designed to fill coverage gaps in existing benchmarks across underrepresented CWE categories. Evaluated on a standardized C vulnerability detection benchmark, VULNSCOUT-C outperforms all evaluated baselines, including state-of-the-art reasoning LLMs and commercial static analysis tools, while offering a fraction of their inference cost. These results demonstrate that task-specialized compact architectures can match or even outperform the detection capability of models orders of magnitude larger, making continuous, low-latency vulnerability analysis practical within real-world development workflows.

## Recorded benchmark results

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

- [Vulnerability Detection on CASTLE 250-sample benchmark 1.0](<https://www.sota2.com/research/sota/vulnerability-detection-on-castle-250-sample-benchmark-1-0.md>): Vulnerability Detection; CASTLE 250-sample benchmark 1.0; CASTLE Score: 1068.

## Original sources and implementation

[Original paper](<https://arxiv.org/abs/2603.28309>)

[PDF](<https://arxiv.org/pdf/2603.28309.pdf>)