paper-with-me

홈 › Papers

Jailbreaking LLMs with Arabic Transliteration and Arabizi

2024-06-26 · Mansour Al Ghanim, Saleh Almohaimeed, Mengxin Zheng, Yan Solihin, Qian Lou

This study identifies the potential vulnerabilities of Large Language Models (LLMs) to 'jailbreak' attacks, specifically focusing on the Arabic language and its various forms. While most research has concentrated on English-based prompt manipulation, our investigation broadens the scope to investigate the Arabic language. We initially tested the AdvBench benchmark in Standardized Arabic, finding that even with prompt manipulation techniques like prefix injection, it was insufficient to provoke LLMs into generating unsafe content. However, when using Arabic transliteration and chatspeak (or arabizi), we found that unsafe content could be produced on platforms like OpenAI GPT-4 and Anthropic Claude 3 Sonnet. Our findings suggest that using Arabic and its various forms could expose information that might remain hidden, potentially increasing the risk of jailbreak attacks. We hypothesize that this exposure could be due to the model's learned connection to specific words, highlighting the need for more comprehensive safety training across all language forms.

📄 PDF Abstract BibTeX arXiv:2406.18725

Code (1)

securedl/arabic_jailbreak 공식 구현 pytorch

Tasks

Transliteration

Methods 이 논문이 사용한 방법론

Attention 설명 없음
Softmax The Softmax output function transforms a previous layer's output into a vector of probabilities. It is commonly used for multiclass classification. Given an input vector $x$…
Layer Normalization Unlike batch normalization, Layer Normalization directly estimates the normalization statistics from the summed inputs…
Absolute Position Encodings Absolute Position Encodings are a type of position embeddings for [Transformer-based models] where positional encodings are…
BPE Byte Pair Encoding, or BPE, is a subword segmentation algorithm that encodes rare and unknown words as sequences of subword units. The intuition is that various word…
Label Smoothing Label Smoothing is a regularization technique that introduces noise for the labels. This accounts for the fact that datasets may have mistakes in them, so maximizing the…
Position-Wise Feed-Forward Layer 설명 없음
Dropout Dropout is a regularization technique for neural networks that drops a unit (along with connections) at training time with a specified probability $p$ (a common value is…

Similar Papers 제목 키워드 기반

Hybrid approach for transliteration of Algerian arabizi: a primary study

2018-08-10 · Imane Guellil, Faical Azouaou, Fodil Benali, Ala-eddine Hachani 외

A hybrid approach for the transliteration of Algerian Arabizi: A primary study In this paper, we present a hybrid approach for the transliteration of the Algerian Arabizi. We define a set of rules enable us the passage f…

Transliteration

A Unified Model for Arabizi Detection and Transliteration using Sequence-to-Sequence Models

2020-12-01 · COLING (WANLP) 2020 12 · Ali Shazal, Aiza Usman, Nizar Habash

While online Arabic is primarily written using the Arabic script, a Roman-script variety called Arabizi is often seen on social media. Although this representation captures the phonology of the language, it is not a one-…

Transliteration

Transliteration of Arabizi into Arabic Orthography: Developing a Parallel Annotated Arabizi-Arabic Script SMS/Chat Corpus

2014-10-01 · WS 2014 10 · Ann Bies, Zhiyi Song, Mohamed Maamouri, Stephen Grimes 외
Transliteration

A Simple but Effective Approach to Improve Arabizi-to-English Statistical Machine Translation

2016-12-01 · WS 2016 12 · Marlies van der Wees, Arianna Bisazza, Christof Monz

A major challenge for statistical machine translation (SMT) of Arabic-to-English user-generated text is the prevalence of text written in Arabizi, or Romanized Arabic. When facing such texts, a translation system trained…

Machine TranslationTranslationTransliteration

Arabizi Detection and Conversion to Arabic

2013-06-28 · WS 2014 10 · Kareem Darwish

Arabizi is Arabic text that is written using Latin characters. Arabizi is used to present both Modern Standard Arabic (MSA) or Arabic dialects. It is commonly used in informal settings such as social networking sites and…

Language ModelingLanguage ModellingTransliteration