Skip to content
GCC AI Research

Testing LLMs safety in Arabic from two perspectives | NAACL

MBZUAI · Notable

Summary

Researchers at MBZUAI presented a new Arabic dataset at NAACL to measure LLM safety, building on a Chinese dataset called 'Do Not Answer'. The dataset includes nearly 5,800 questions with challenges and harmless requests containing sensitive terms to test for over-sensitivity. The team localized cultural concepts and added 3,000 questions specific to Arabic language and culture. Why it matters: This comprehensive benchmark, accounting for the diversity of Arabic dialects and cultures, advances the development of safer and more culturally aligned LLMs for Arabic speakers.

Keywords

LLM safety · Arabic · MBZUAI · NAACL · dataset

Get the weekly digest

Top AI stories from the GCC region, every week.

Related

SalamahBench: Toward Standardized Safety Evaluation for Arabic Language Models

arXiv ·

The paper introduces SalamahBench, a new benchmark for evaluating the safety of Arabic Language Models (ALMs). The benchmark comprises 8,170 prompts across 12 categories aligned with the MLCommons Safety Hazard Taxonomy. Five state-of-the-art ALMs, including Fanar 1 and 2, ALLaM 2, Falcon H1R, and Jais 2, were evaluated using the benchmark. Why it matters: The benchmark enables standardized, category-aware safety evaluation, highlighting the necessity of specialized safeguard mechanisms for robust harm mitigation in ALMs.

AraTrust: An Evaluation of Trustworthiness for LLMs in Arabic

arXiv ·

The paper introduces AraTrust, a new benchmark for evaluating the trustworthiness of LLMs when prompted in Arabic. The benchmark contains 522 multiple-choice questions covering dimensions like truthfulness, ethics, safety, and fairness. Experiments using AraTrust showed that GPT-4 performed the best, while open-source models like AceGPT 7B and Jais 13B had lower scores. Why it matters: This benchmark addresses a critical gap in evaluating LLMs for Arabic, which is essential for ensuring the safe and ethical deployment of AI in the Arab world.

Evaluating Arabic Large Language Models: A Survey of Benchmarks, Methods, and Gaps

arXiv ·

This survey paper analyzes over 40 benchmarks used to evaluate Arabic large language models, categorizing them into Knowledge, NLP Tasks, Culture and Dialects, and Target-Specific evaluations. It identifies progress in benchmark diversity but also highlights gaps like limited temporal evaluation and cultural misalignment. The paper also examines methods for creating benchmarks, including native collection, translation, and synthetic generation. Why it matters: The survey provides a comprehensive reference for Arabic NLP research and offers recommendations for future benchmark development to better align with cultural contexts.