الملخص
Retrieval-Augmented Generation (RAG) systems are vulnerable to prompt-injection attacks embedded in retrieved content. We introduce RAG-PIBench, a benchmark for RAG-style prompt-injection detection containing 4,876 contextual examples across frozen train, validation, and protected-test splits. Using a leakage-aware construction pipeline and strict evaluation protocol, we compare keyword-based, semantic-reference, TF-IDF, and transformer-based detectors. DistilBERT achieves the best protected-test performance (F1 = 0.896, PR-AUC = 0.968), while TF-IDF SVM and logistic regression remain competitive. Our results demonstrate the value of leakage-aware benchmark design and strong sparse baselines for reliable prompt-injection detection in RAG systems.
الكلمات المفتاحية
الموضوع
بيانات النشر
- المجلة
- غير متاح
- وصول مفتوح
- وصول مفتوح أخضر
اقتبس هذه المقالة
APA 7
Jaffal, N. O., Yuksel, A., & Mohaisen, D. (2026). RAG-PIBench: A Leakage-Aware Benchmark for Prompt-Injection Detection in Trustworthy RAG Systems. https://omanscience.com/ar/articles/rag-pibench-a-leakage-aware-benchmark-for-prompt-injection-detection-in-trustworthy-rag-systems
MLA 9
Jaffal, Niveen O., et al. "RAG-PIBench: A Leakage-Aware Benchmark for Prompt-Injection Detection in Trustworthy RAG Systems." https://omanscience.com/ar/articles/rag-pibench-a-leakage-aware-benchmark-for-prompt-injection-detection-in-trustworthy-rag-systems.
شيكاغو (المؤلف–التاريخ)
Jaffal, Niveen O., Ahmet Yuksel, and David Mohaisen. 2026. "RAG-PIBench: A Leakage-Aware Benchmark for Prompt-Injection Detection in Trustworthy RAG Systems." https://omanscience.com/ar/articles/rag-pibench-a-leakage-aware-benchmark-for-prompt-injection-detection-in-trustworthy-rag-systems.
هارفارد
Jaffal, N. O., Yuksel, A. and Mohaisen, D. (2026) 'RAG-PIBench: A Leakage-Aware Benchmark for Prompt-Injection Detection in Trustworthy RAG Systems', Available at: https://omanscience.com/ar/articles/rag-pibench-a-leakage-aware-benchmark-for-prompt-injection-detection-in-trustworthy-rag-systems.
فانكوفر
Jaffal NO, Yuksel A, Mohaisen D. RAG-PIBench: A Leakage-Aware Benchmark for Prompt-Injection Detection in Trustworthy RAG Systems. https://omanscience.com/ar/articles/rag-pibench-a-leakage-aware-benchmark-for-prompt-injection-detection-in-trustworthy-rag-systems
IEEE
N. O. Jaffal, A. Yuksel, and D. Mohaisen, "RAG-PIBench: A Leakage-Aware Benchmark for Prompt-Injection Detection in Trustworthy RAG Systems," https://omanscience.com/ar/articles/rag-pibench-a-leakage-aware-benchmark-for-prompt-injection-detection-in-trustworthy-rag-systems.