← all datasets

English and multilingual prompt-safety benchmarks

Emerging
1papers using it
2026first seen

The 'English and multilingual prompt-safety benchmarks' consist of a set of evaluations designed to assess the safety performance of AI models across various prompts in both English and multiple other languages.

🔎 Find this dataset

Papers using English and multilingual prompt-safety benchmarks (1)

English and multilingual prompt-safety benchmarks — datasets — cybersecurity