Research

Paper

AI LLM March 11, 2026

TOSSS: a CVE-based Software Security Benchmark for Large Language Models

Authors

Marc Damie, Murat Bilgehan Ertan, Domenico Essoussi, Angela Makhanu, Gaëtan Peter, Roos Wensveen

Abstract

With their increasing capabilities, Large Language Models (LLMs) are now used across many industries. They have become useful tools for software engineers and support a wide range of development tasks. As LLMs are increasingly used in software development workflows, a critical question arises: are LLMs good at software security? At the same time, organizations worldwide invest heavily in cybersecurity to reduce exposure to disruptive attacks. The integration of LLMs into software engineering workflows may introduce new vulnerabilities and weaken existing security efforts. We introduce TOSSS (Two-Option Secure Snippet Selection), a benchmark that measures the ability of LLMs to choose between secure and vulnerable code snippets. Existing security benchmarks for LLMs cover only a limited range of vulnerabilities. In contrast, TOSSS relies on the CVE database and provides an extensible framework that can integrate newly disclosed vulnerabilities over time. Our benchmark gives each model a security score between 0 and 1 based on its behavior; a score of 1 indicates that the model always selects the secure snippet, while a score of 0 indicates that it always selects the vulnerable one. We evaluate 14 widely used open-source and closed-source models on C/C++ and Java code and observe scores ranging from 0.48 to 0.89. LLM providers already publish many benchmark scores for their models, and TOSSS could become a complementary security-focused score to include in these reports.

Metadata

arXiv ID: 2603.10969
Provider: ARXIV
Primary Category: cs.LG
Published: 2026-03-11
Fetched: 2026-03-12 04:21

Related papers

Raw Data (Debug)
{
  "raw_xml": "<entry>\n    <id>http://arxiv.org/abs/2603.10969v1</id>\n    <title>TOSSS: a CVE-based Software Security Benchmark for Large Language Models</title>\n    <updated>2026-03-11T16:54:01Z</updated>\n    <link href='https://arxiv.org/abs/2603.10969v1' rel='alternate' type='text/html'/>\n    <link href='https://arxiv.org/pdf/2603.10969v1' rel='related' title='pdf' type='application/pdf'/>\n    <summary>With their increasing capabilities, Large Language Models (LLMs) are now used across many industries. They have become useful tools for software engineers and support a wide range of development tasks. As LLMs are increasingly used in software development workflows, a critical question arises: are LLMs good at software security? At the same time, organizations worldwide invest heavily in cybersecurity to reduce exposure to disruptive attacks. The integration of LLMs into software engineering workflows may introduce new vulnerabilities and weaken existing security efforts.\n  We introduce TOSSS (Two-Option Secure Snippet Selection), a benchmark that measures the ability of LLMs to choose between secure and vulnerable code snippets. Existing security benchmarks for LLMs cover only a limited range of vulnerabilities. In contrast, TOSSS relies on the CVE database and provides an extensible framework that can integrate newly disclosed vulnerabilities over time. Our benchmark gives each model a security score between 0 and 1 based on its behavior; a score of 1 indicates that the model always selects the secure snippet, while a score of 0 indicates that it always selects the vulnerable one. We evaluate 14 widely used open-source and closed-source models on C/C++ and Java code and observe scores ranging from 0.48 to 0.89. LLM providers already publish many benchmark scores for their models, and TOSSS could become a complementary security-focused score to include in these reports.</summary>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.LG'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CL'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.CR'/>\n    <category scheme='http://arxiv.org/schemas/atom' term='cs.SE'/>\n    <published>2026-03-11T16:54:01Z</published>\n    <arxiv:primary_category term='cs.LG'/>\n    <author>\n      <name>Marc Damie</name>\n    </author>\n    <author>\n      <name>Murat Bilgehan Ertan</name>\n    </author>\n    <author>\n      <name>Domenico Essoussi</name>\n    </author>\n    <author>\n      <name>Angela Makhanu</name>\n    </author>\n    <author>\n      <name>Gaëtan Peter</name>\n    </author>\n    <author>\n      <name>Roos Wensveen</name>\n    </author>\n  </entry>"
}