From 96028ef3c33f0c19a59cb0bfd45ae949e5805183 Mon Sep 17 00:00:00 2001 From: Francesco Balassone <77972200+duel0@users.noreply.github.com> Date: Mon, 27 Oct 2025 10:42:07 +0100 Subject: [PATCH] Added titles --- benchmarks/README.md | 22 ++++++++++++++++++---- 1 file changed, 18 insertions(+), 4 deletions(-) diff --git a/benchmarks/README.md b/benchmarks/README.md index 0fae3d16..0689760a 100644 --- a/benchmarks/README.md +++ b/benchmarks/README.md @@ -28,10 +28,24 @@ AutoPenBench │ *Categories marked with asterisk are available in CAI PRO version [^8]. - -| ![](utils/stackplot.png) | ![](utils/base_1col.png) | -|------------------------------------------------|---------------------------------| -| ![](utils/cyberpii_benchmark.png) | ![](utils/caibench_spider.png) | + + + + + + + + + + + + + + + + + +
Best performance in Agent vs Agent A&DModel performance in Jeopardy CTFs Base Benchmark
stackplotbase_1col
Model performance in CyberPII Privacy BenchmarkModel performance overall
cyberpiicaibench
Cybersecurity AI Benchmark or `CAIBench` for short is a meta-benchmark (*benchmark of benchmarks*) [^6] designed to evaluate the security capabilities (both offensive and defensive) of cybersecurity AI agents and their associated models. It is built as a composition of individual benchmarks, most represented by a Docker container for reproducibility. Each container scenario can contain multiple challenges or tasks. The system is designed to be modular and extensible, allowing for the addition of new benchmarks and challenges.