AIxCC 2025 탐지율 90% 달성 및 Zero-day 29개 발굴한 Multi-Agent 보안 시스템
Multi-Agent LLM System for Automated Vulnerability Discovery and Reproduction
Multi-Agent LLM System for Automated Vulnerability Discovery and Reproduction
Anthropic to release Mythos-class models to the public
PREDICTION-20260518-0006: craft-and-peer-recognition [2026-Q3 through 2027-Q4]
Anthropic's Project Glasswing CVE tally is still anyone's guess
N-Day-Bench – Can LLMs find real vulnerabilities in real codebases?
Evaluation of Claude Mythos Preview's cyber capabilities