license: apache-2.0
task_categories:
- text-classification
language: - ar
- en
pretty_name: ArabGuard-v1
size_categories: - 1K-10K
tags: - safety
- prompt-injection
- egyptian-dialect
- cybersecurity
- jailbreak
π‘οΈ ArabGuard-v1: Adversarial Dialectal Dataset for LLM Safety
ArabGuard-v1 is a specialized, manually curated benchmark designed to evaluate and enhance the robustness of Large Language Models (LLMs) against Prompt Injection and Jailbreaking attacks, specifically focusing on the Egyptian Dialect and Franco-Arabic.
π Overview
Global safety guardrails often exhibit a "Linguistic Blind Spot" when faced with local cultural nuances, slang, or code-switching. ArabGuard-v1 bridges this gap by providing 2,321 unique samples of adversarial prompts that leverage social engineering, roleplay, and complex logical manipulation (including Gaslighting) in a localized Arabic context.
π Dataset Statistics & Splits
The dataset has been rigorously deduplicated (715 duplicates removed) and partitioned using a stratified split to ensure consistent label distribution across all sets:
| Split | Samples | Malicious (1) | Benign (0) |
|---|---|---|---|
| Train | 1,856 | ~1,014 | ~842 |
| Validation | 232 | ~127 | ~105 |
| Test | 233 | ~127 | ~106 |
| Total (Unique) | 2,321 | 1,268 | 1,053 |
π Attack Taxonomy
The dataset covers sophisticated attack vectors tailored to the MENA region:
- Linguistic Obfuscation: Payload splitting (e.g., Ψͺ-Ψ¬-Ψ§-Ω-Ω) and Franco-Arabic.
- Gaslighting Traps: Repetitive correction and negative reinforcement to force model compliance.
- Social Engineering: Leveraging local idioms and authoritative impersonation.
- Hard Negatives: Benign technical queries (e.g., "How to prevent SQL Injection") to minimize over-refusal.
ποΈ Dataset Structure
text: The raw input prompt.attack_type: Category (e.g.,gaslighting_trap,prompt_leaking,reverse_psychology).dialect: Linguistic style (egyptian_slang,franco,msa_mixed).intent_goal: The ultimate objective (e.g.,exfiltration,manipulation).label: Binary (1: Malicious, 0: Safe).
π οΈ Usage
To use this dataset with the datasets library:
from datasets import load_dataset
dataset = load_dataset("d12o6aa/ArabGuard-Egyptian-V1")
π License
This dataset is licensed under the Apache License 2.0.