# ZeroDayBench: Evaluating LLM Agents on Unseen Zero-Day Vulnerabilities for Cyberdefense

arXiv: https://arxiv.org/abs/2603.02297 · HTML: https://mlsanity.com/paper/2603.02297/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Placed in Agent threats, guardrails and governance, one of 16 directions in LLM agents and tool use.
- Builds on 15 of its 25 references that are in the map.
- First posted on arXiv 2 Mar 2026.

## Placed in

- LLM agents and tool use / [Agent threats, guardrails and governance](https://mlsanity.com/agents/agent-threats-guardrails-and-governance/)

## Builds on (most cited first, 15 of 15)

- [SWE-bench: Can Language Models Resolve Real-World GitHub Issues?](https://mlsanity.com/paper/2310.06770/) (2023)
- [SWE-agent: Agent-Computer Interfaces Enable Automated Software Engineering](https://mlsanity.com/paper/2405.15793/) (2024)
- [A Survey on Large Language Model (LLM) Security and Privacy: The Good, the Bad, and the Ugly](https://mlsanity.com/paper/2312.02003/) (2023)
- [CVE-Bench: A Benchmark for AI Agents' Ability to Exploit Real-World Web Application Vulnerabilities](https://mlsanity.com/paper/2503.17332/) (2025)
- [CyberGym: Evaluating AI Agents' Real-World Cybersecurity Capabilities at Scale](https://mlsanity.com/paper/2506.02548/) (2025)
- [SEC-bench: Automated Benchmarking of LLM Agents on Real-World Software Security Tasks](https://mlsanity.com/paper/2506.11791/) (2025)
- [Securing Large Language Models: Threats, Vulnerabilities and Responsible Practices](https://mlsanity.com/paper/2403.12503/) (2024)
- [Benchmarking LLMs and LLM-based Agents in Practical Vulnerability Detection for Code Repositories](https://mlsanity.com/paper/2503.03586/) (2025)
- [PATCHEVAL: A New Benchmark for Evaluating LLMs on Patching Real-World Vulnerabilities](https://mlsanity.com/paper/2511.11019/) (2025)
- [SecVulEval: Benchmarking LLMs for Real-World C/C++ Vulnerability Detection](https://mlsanity.com/paper/2505.19828/) (2025)
- [Benchmarking Practices in LLM-driven Offensive Security: Testbeds, Metrics, and Experiment Design](https://mlsanity.com/paper/2504.10112/) (2025)
- [VulnRepairEval: An Exploit-Based Evaluation Framework for Assessing Large Language Model Vulnerability Repair Capabilities](https://mlsanity.com/paper/2509.03331/) (2025)
- [Cybersecurity AI Benchmark (CAIBench): A Meta-Benchmark for Evaluating Cybersecurity AI Agents](https://mlsanity.com/paper/2510.24317/) (2025)
- [VADER: A Human-Evaluated Benchmark for Vulnerability Assessment, Detection, Explanation, and Remediation](https://mlsanity.com/paper/2505.19395/) (2025)
- [StriderSPD: Structure-Guided Joint Representation Learning for Binary Security Patch Detection](https://mlsanity.com/paper/2601.05772/) (2026)

## Built on by (most cited first, 4 of 4)

- [SEC-bench Pro: Can Language Models Solve Long-Horizon Software Security Tasks?](https://mlsanity.com/paper/2605.26548/) (2026)
- [Are Frontier LLMs Ready for Cybersecurity? Evidence for Vertical Foundation Models from Dual-Mode Vulnerability Benchmarks](https://mlsanity.com/paper/2605.23243/) (2026)
- [FORGE: Multi-Agent Graduated Exploitation and Detection Engineering](https://mlsanity.com/paper/2606.03453/) (2026)
- [CWE-bench: Refocusing Agentic Security from Exploitation to Defense](https://mlsanity.com/paper/2610.17020/) (2026)

## Abstract (arXiv)

Large language models (LLMs) are increasingly being deployed as software engineering agents that autonomously contribute to repositories. A major benefit these agents present is their ability to find and patch security vulnerabilities in the codebases they oversee. To estimate the capability of agents in this domain, we introduce ZeroDayBench, a benchmark where LLM agents find and patch 22 novel critical vulnerabilities in open-source codebases. We focus our efforts on three popular frontier agentic LLMs: GPT-5.2, Claude Sonnet 4.5, and Grok 4.1. We find that frontier LLMs are not yet capable of autonomously solving our tasks and observe some behavioral patterns that suggest how these models can be improved in the domain of proactive cyberdefense.