Lucene search
+L

When Models Know but Can't Linearly Say: Relational Encoding Limits Linear Vulnerability Classification in Code LLMs

🗓️ 19 Oct 2026 00:00:00Reported by Rui Abreu, Corina Pasareanu, Claudia Mamede, Andre Catarino, Rui MeloType 
packetstormnews
 packetstormnews
🔗 packetstorm.news👁 1 Views

Code LLMs encode vulnerability as relational signals rather than separable categorical labels.

Data

Build on a solid foundation with Vulners data

We provide the essential building blocks for cybersecurity solutions with comprehensive, structured, and constantly updated vulnerability and exploits data

Api

Power your application with Vulners API

The Vulners REST API offers reliable, high-performance access to vulnerability intelligence, with 99.9% SLA uptime and CDN-backed data delivery for seamless global access

App

Assess and manage vulnerabilities with Vulners tools

Built on top of Vulners' database and SDK, end-user solutions give security professionals and developers lightweight and powerful tools for vulnerability remediation

19 Oct 2026 00:00Current
5.9Medium risk
Vulners AI Score5.9
1