# HoF-Bench: Rediscovering Real AI-Discovered CVEs Without Frontier Models

arXiv: https://arxiv.org/abs/2607.27030 · HTML: https://mlsanity.com/paper/2607.27030/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 9 of its 22 references that are in the map.
- First posted on arXiv 29 Jul 2026.

## Builds on (most cited first, 9 of 9)

- [Devign: Effective Vulnerability Identification by Learning Comprehensive Program Semantics via Graph Neural Networks](https://mlsanity.com/paper/1909.03496/) (2019)
- [Vulnerability Detection with Code Language Models: How Far Are We?](https://mlsanity.com/paper/2403.18624/) (2024)
- [Practical Bayesian model evaluation using leave-one-out cross-validation and WAIC](https://mlsanity.com/paper/1507.04544/) (2015)
- [SEC-bench: Automated Benchmarking of LLM Agents on Real-World Software Security Tasks](https://mlsanity.com/paper/2506.11791/) (2025)
- [BountyBench: Dollar Impact of AI Agent Attackers and Defenders on Real-World Cybersecurity Systems](https://mlsanity.com/paper/2505.15216/) (2025)
- [Benchmarking LLMs and LLM-based Agents in Practical Vulnerability Detection for Code Repositories](https://mlsanity.com/paper/2503.03586/) (2025)
- [RealVuln: Benchmarking Rule-Based, General-Purpose LLM, and Security-Specialized Scanners on Real-World Code](https://mlsanity.com/paper/2604.13764/) (2026)
- [DREA: Decoupled Reasoning and Exploration Agents for Repository-Level Vulnerability Detection](https://mlsanity.com/paper/2607.13439/) (2026)
- [Snyk VulnBench JS 1.0: Can LLMs Find the Same Bugs Twice?](https://mlsanity.com/paper/2606.15762/) (2026)

## Abstract (arXiv)

LLM-based analyzers have begun finding real vulnerabilities in mature open-source projects: AISLE's analyzer is credited with more than 280 CVEs across 78 projects, including OpenSSL, curl, and GnuTLS. We introduce HoF-Bench (named after AISLE's public Hall of Fame), a benchmark built from 95 of these public AI-discovered CVEs across eight repositories pinned at vulnerable commits. Analyzers receive source and target-file scope but not CVE identifiers, descriptions, fixes, or expected mechanisms; a detector-blinded frontier-model judge credits only findings that identify the same code path, root cause, attack condition, and impact. A deliberately minimal LLM-based analyzer rediscovers up to 65 of the 95 CVEs (68%) under this strict protocol. No frontier model performs detection anywhere in the study. The ten detector backbones are five open-weight models (21B--284B total parameters, 3--13B active) and five proprietary small or "flash"-tier models. All of them run in the fixed scaffold with four repeated passes, an optional generated-context stage, and a replayable multi-round triage stage (7,600 model--CVE pass records). Difficulty is strongly structured by language; the CVEs missed by every model concentrate in C infrastructure code. HoF-Bench provides a compact test bed for comparing vulnerability scanners, their reliability across repeated runs, and the candidate volume they create. The dataset is available at https://huggingface.co/datasets/aisleinc/HoF-Bench.