0

FORTRESS

Active

A dataset of 500 expert-crafted adversarial prompts with instance-based rubrics of 4-7 binary questions for automated evaluation across 3 domains relevant to national security and public safety (NSPS).

Publisher
Scale AI
Domain
Safeguards
License
mit
Published
Oct 2025
Notable for
Benchmark for evaluating Safeguards.

Cite

Notes

Only stored in your browser.

FAQ

What is FORTRESS?
A dataset of 500 expert-crafted adversarial prompts with instance-based rubrics of 4-7 binary questions for automated evaluation across 3 domains relevant to national security and public safety (NSPS).
What license is FORTRESS under?
FORTRESS is available under mit.