HalluScore is a curated Arabic QA dataset with 827 questions, ground-truth evidence, and human annotations used to measure hallucination rates across 17 LLMs.
Jais and jais-chat: Arabic-centric foundation and instruction-tuned open gener- ative large language models
10 Pith papers cite this work. Polarity classification is still indexing.
10
Pith papers citing it
fields
cs.CL 10representative citing papers
XL-SafetyBench is a new cross-cultural benchmark showing frontier LLMs decouple jailbreak robustness from cultural sensitivity while local models trade off attack success against neutral-safe rates in a near-linear pattern indicating generation failure rather than alignment.
LQM introduces a six-level linguistically motivated error taxonomy for MT evaluation and applies it via expert annotation to LLM outputs on a new 3,850-sentence multi-dialect Arabic corpus.
IndicGuard provides a culturally nuanced safety dataset for ten Indic languages and a fine-tuned Gemma-3-4B-IT model that outperforms CultureGuard on moderation tasks and generalizes to unseen low-resource languages.
ArabiGEE introduces the first hierarchical taxonomy for Arabic grammatical error explanation, with 27 error types, 140 correction types, and 324 explanations, applied to annotate corpora and evaluate LLMs.
Using a 1PL IRT model on real cultural questions across 13 locales, the study identifies a local-language knowledge-access advantage masked by lower proficiency in raw accuracy.
ArabCulture-Dialogue dataset shows LLMs perform worse on dialectal Arabic than Modern Standard Arabic across cultural reasoning, translation, and generation tasks.
Translating unsafe inputs to low-resource languages jailbreaks GPT-4 at rates on par with or exceeding state-of-the-art attacks.
Introduces a multi-stage Arabic financial sentiment pipeline that produces an 84K-sample corpus for company-level analysis tied to Saudi stock market behavior.
Residual-stream noise injection raises narrative diversity in Arabic educational stories while preserving reading-grade level, outperforming high-temperature sampling across five 7-9B models.
citing papers explorer
-
HalluScore: Large Language Model Hallucination Question Answering Benchmark
HalluScore is a curated Arabic QA dataset with 827 questions, ground-truth evidence, and human annotations used to measure hallucination rates across 17 LLMs.
-
XL-SafetyBench: A Country-Grounded Cross-Cultural Benchmark for LLM Safety and Cultural Sensitivity
XL-SafetyBench is a new cross-cultural benchmark showing frontier LLMs decouple jailbreak robustness from cultural sensitivity while local models trade off attack success against neutral-safe rates in a near-linear pattern indicating generation failure rather than alignment.
-
LQM: Linguistically Motivated Multidimensional Quality Metrics for Machine Translation
LQM introduces a six-level linguistically motivated error taxonomy for MT evaluation and applies it via expert annotation to LLM outputs on a new 3,850-sentence multi-dialect Arabic corpus.
-
IndicGuard: A Multilingual Safety Guard Model and Dataset for Indic Languages
IndicGuard provides a culturally nuanced safety dataset for ten Indic languages and a fine-tuned Gemma-3-4B-IT model that outperforms CultureGuard on moderation tasks and generalizes to unseen low-resource languages.
-
ArabiGEE: A Hierarchical Taxonomy for Arabic Grammatical Error Explanation
ArabiGEE introduces the first hierarchical taxonomy for Arabic grammatical error explanation, with 27 error types, 140 correction types, and 324 explanations, applied to annotate corpora and evaluate LLMs.
-
The Masked Advantage: Uncovering Local-Language Access to Cultural Knowledge in LLMs
Using a 1PL IRT model on real cultural questions across 13 locales, the study identifies a local-language knowledge-access advantage masked by lower proficiency in raw accuracy.
-
Cultural Benchmarking of LLMs in Standard and Dialectal Arabic Dialogues
ArabCulture-Dialogue dataset shows LLMs perform worse on dialectal Arabic than Modern Standard Arabic across cultural reasoning, translation, and generation tasks.
-
LLM-Based Financial Sentiment Analysis in Arabic: Evidence from Saudi Markets
Introduces a multi-stage Arabic financial sentiment pipeline that produces an 84K-sample corpus for company-level analysis tied to Saudi stock market behavior.
-
Noise Steering for Controlled Text Generation: Improving Diversity and Reading-Level Fidelity in Arabic Educational Story Generation
Residual-stream noise injection raises narrative diversity in Arabic educational stories while preserving reading-grade level, outperforming high-temperature sampling across five 7-9B models.