We introduce JEEM, a benchmark designed to evaluate Vision-Language Models (VLMs) on visual understanding across four Arabic-speaking countries: Jordan, The Emirates, Egypt, and Morocco. JEEM includes the tasks of image captioning and visual question answering, and features culturally rich and regionally diverse content. This dataset aims to assess the ability of VLMs to generalize across dialects and accurately interpret cultural elements in visual contexts. In an evaluation of five prominent open-source Arabic VLMs and GPT-4o, we find that the Arabic VLMs consistently underperform, struggling with both visual understanding and dialect-specific generation. While GPT-4o ranks best in this comparison, the model’s linguistic competence varies across dialects, and its visual understanding capabilities lag behind. This underscores the need for more inclusive models and the value of culturally-diverse evaluation paradigms.

Speech foundation models trained at a massive scale, both in terms of model and data size, result in robust systems capable of performing multiple speech tasks, including automatic speech recognition (ASR). These models transcend language and domain barriers, yet effectively measuring their performance remains a challenge. Traditional metrics like word error rate (WER) and character error rate (CER) are commonly used to evaluate ASR performance but often fail to reflect transcription quality in critical contexts, particularly when detecting fabricated outputs. This phenomenon, known as hallucination, is especially concerning in high-stakes domains such as healthcare, legal, and aviation, where errors can have severe consequences. In our work, we address this gap by investigating hallucination in ASR models. We examine how factors such as distribution shifts, model size, and model architecture influence the hallucination error rate (HER), a metric we introduce to quantify hallucinations. Our analysis of over 20 ASR models reveals key insights: (1) High WERs can mask low hallucination rates, while low WERs may conceal dangerous hallucinations. (2) Synthetic noise, both adversarial and common perturbations like white noise, pitch shift, and time stretching, increase HER. (3) Distribution shift correlates strongly with HER (𝛼 = 0.91). Our findings highlight the importance of incorporating HER alongside traditional metrics like WER to better assess ASR model performance, particularly in high-stakes domains.

Recent advances in speech foundation models are largely driven by scaling both model size and data, enabling them to perform a wide range of tasks, including speech recognition. Traditionally, ASR models are evaluated using metrics like Word Error Rate (WER) and Character Error Rate (CER), which depend on ground truth labels. As a result of limited labeled data from diverse domains and testing conditions, the true generalization capabilities of these models beyond standard benchmarks remain unclear. Moreover, labeling data is both costly and time-consuming. To address this, we propose a novel label-free approach for approximating ASR performance metrics, eliminating the need for ground truth labels. Our method utilizes multimodal embeddings in a unified space for speech and transcription representations, combined with a high-quality proxy model to compute proxy metrics. These features are used to train a regression model to predict key ASR metrics like Word Error Rate (WER) and Character Error Rate (CER). We experiment with over 40 models across 14 datasets representing both standard and in-the-wild testing conditions. Our results show that we approximate the metrics within a single-digit absolute difference across all experimental configurations, outperforming the most recent baseline by more than 50%.

As large language models (LLMs) become increasingly integrated into daily life, ensuring their cultural sensitivity and inclusivity is paramount. We introduce PALM, a year-long community-driven project covering all 22 Arab countries. The dataset contains instruction–response pairs in both Modern Standard Arabic (MSA) and dialectal Arabic (DA), spanning 20 diverse topics. Built by a team of 44 researchers across the Arab world—each an author of this paper—PALM offers a broad, inclusive perspective. We use PALM to evaluate the cultural and dialectal capabilities of several frontier LLMs, revealing notable limitations: while closed-source LLMs generally perform strongly, they still exhibit flaws, and smaller open-source models face greater challenges. Furthermore, certain countries (e.g., Egypt, the UAE) appear better represented than others (e.g., Iraq, Mauritania, Yemen). Our annotation guidelines, code, and data are publicly available for reproducibility. More information about PALM is available on our project page: https://github.com/UBC-NLP/palm.

Large language models (LLMs) play a crucial role in a wide range of real world applications. However, concerns about their safety and ethical implications are growing. While research on LLM safety is expanding, there is a noticeable gap in evaluating safety across multiple languages, especially in Arabic and Russian. We address this gap by exploring biases in LLMs across different languages and contexts, focusing on GPT-3.5 and Gemini. Through carefully designed argument-based prompts and scenarios in Arabic, English, and Russian, we examine biases in cultural, political, racial, religious, and gender domains. Our findings reveal biases in these domains. In particular, our investigation uncovers subtle biases where each model tends to present winners as those speaking the primary language the model is prompted with. Our study contributes to ongoing efforts to ensure justice and equality in LLM development and emphasizes the importance of further research towards responsible progress in this field.

We present the results of Shared Task “Dialect to MSA Translation”, which tackles challenges posed by the diverse Arabic dialects in machine translation. Covering Gulf, Egyptian, Levantine, Iraqi and Maghrebi dialects, the task offers 1001 sentences in both MSA and dialects for fine-tuning, alongside 1888 blind test sentences. Leveraging GPT-3.5, a state-of-the-art language model, our method achieved the a BLEU score of 29.61. This endeavor holds significant implications for Neural Machine Translation (NMT) systems targeting low-resource langu ages with linguistic variation. Additionally, negative experiments involving fine-tuning AraT5 and No Language Left Behind (NLLB) using the MADAR Dataset resulted in BLEU scores of 10.41 and 11.96, respectively. Future directions include expanding the dataset to incorporate more Arabic dialects and exploring alternative NMT architectures to further enhance translation capabilities.

Navigating the intricacies of machine translation (MT) involves tackling the nuanced disparities between Arabic dialects and Modern Standard Arabic (MSA), presenting a formidable obstacle. In this study, we delve into Subtask 3 of the NADI shared task (CITATION), focusing on the translation of sentences from four distinct Arabic dialects into MSA. Our investigation explores the efficacy of various models, including Jais, NLLB, GPT-3.5, and GPT-4, in this dialect-to-MSA translation endeavor. Our findings reveal that Jais surpasses all other models, boasting an average BLEU score of 19.48 in the combination of zero- and few-shot setting, whereas NLLB exhibits the least favorable performance, garnering a BLEU score of 8.77.