WMDP

Dataset Card for WMDP The Weapons of Mass Destruction Proxy (WMDP) benchmark is a dataset of multiple-choice questions that serve as a proxy measurement of hazardous knowledge in biosecurity, cybersecurity, and chemical security. WMDP serves two roles: first, as an evaluation for hazardous knowledge in LLMs, and second

Papers using WMDP (3)

LLMs Outperform Experts on Challenging Biology Benchmarks2025 · 2 citesCross-Disciplinary Knowledge Retrieval and Synthesis: A Compound AI Architecture for Scientific Discovery2025 · 1 citesEvoMU: Evolutionary Machine Unlearning2026