ExploitBench
Seunghyun Lee · Carnegie Mellon University
[ Abstract ]
Exploitation is a ladder, not a coin flip. ExploitBench grades LLM agents across 16 capability flags against real V8, from first crash to arbitrary code execution, and shows where today's models actually stop.
Benchmarking LLMs on vulnerability discovery and exploitation capabilities — results and lessons learned.
Get the next talks first
New recordings as they land, the call for papers the moment it opens, and the next [un]prompted announced here before anywhere else. Low volume. Unsubscribe anytime.