Advancing Software Vulnerability Detection with Reasoning LLMs: DeepSeek-R1′s Performance and Insights

The increasing complexity of software systems has heightened the need for efficient and accurate vulnerability detection. Large Language Models have emerged as promising tools in this domain; however, their reasoning capabilities and limitations remain insufficiently explored. This study presents a...

ver descrição completa

Detalhes bibliográficos

Principais autores:	Wenting Qin, Lijie Suo, Liangchen Li, Fan Yang
Formato:	Artigo
Idioma:	Inglês
Publicado em:	MDPI AG 2025-06-01
Colecção:	Applied Sciences
Assuntos:	DeepSeek-R1 Large Language Model reasoning mechanisms step-by-step inference vulnerability detection
Acesso em linha:	https://www.mdpi.com/2076-3417/15/12/6651
Tags:	Adicionar Tag Sem tags, seja o primeiro a adicionar uma tag!