The paper introduces FanarGuard, a bilingual moderation filter for Arabic and English language models that considers both safety and cultural alignment. A dataset of 468K prompt-response pairs was created and scored by LLM judges on harmlessness and cultural awareness to train the filter. The first benchmark targeting Arabic cultural contexts was developed to evaluate cultural alignment. Why it matters: FanarGuard advances context-sensitive AI safeguards by integrating cultural awareness into content moderation, addressing a critical gap in current alignment techniques.
The UAE has arrested 35 individuals for publishing misinformation and inciting hatred related to Iran's attacks. These individuals allegedly spread false information and inflammatory content online. The arrests are part of a broader effort to maintain social stability and prevent the spread of harmful content. Why it matters: The arrests highlight the UAE's focus on regulating online discourse and countering narratives perceived as threatening national security or social harmony.
A new content improvement system has been developed to address issues of randomness and incorrectness in text generated by deep learning models like GPT-3. The system uses text mining to identify correct sentences and employs syntactic/semantic generalization to substitute problematic elements. The system can substantially improve the factual correctness and meaningfulness of raw content. Why it matters: Improving the quality of automatically generated content is crucial for ensuring reliability and trustworthiness across various AI applications.
MBZUAI 2023 graduate Muhammad Umar is researching propaganda detection in low-resource, code-switched languages like Roman Urdu. His master's thesis focuses on detecting propaganda techniques in social media text using deep learning models. Umar aims to submit a paper on his findings to the EMNLP 2023 conference. Why it matters: This research addresses the under-explored area of propaganda detection in low-resource languages, which is crucial for combating misinformation in bilingual communities.
This paper presents six experiments evaluating personalization and user tracking in web search engine results. The experiments involve comparing search results based on VPN location (including UAE vs others), logged-in status, network type, search engine, browser, and trained Google accounts. The study measures total hits, first hit, and correlation between hits to identify patterns of personalization. Why it matters: The findings shed light on the extent of filter bubble effects and potential biases in search results for users in the UAE and globally.