Spontaneous Arabic speech is scarce in current corpora, and it is not well represented. This poses a limitation invisibility of spontaneous Arabic to automatic speech recognition (ASR), speaker diarization, and sociolinguistic research. The Saudi ASWAT project fills a major gap by creating the first nationwide corpus of natural Saudi speech, where data has been recorded and transcribed under a systematic methodology and ecologically valid conditions. The corpus aims to collect 2,500 hours of natural conversations from a diverse range of participants. These has been selected from five major Saudi regional varieties, Najdi (Central), Eastern, Hijazi (Western), Northern, and Southern, covering more than fifty five local varieties. Speech has been recorded by trained fieldworkers using participants own devices to reflect real-life variation. The annotated data incorporate a variety of speaker demographics, regional vocabularies which differ from the standard lexicon, and structured metadata. TF–IDF profiling shows regional differences in a range of performing words. Data also represent balanced age and gender sampling to support studies of intergenerational and sociophonetic variation. Saudi ASWAT provides the most linguistically diverse resources of Saudi Arabia to date. Additionally, it establishes an ethical governed framework for Arabic speech data creation to enable advances in both computational modeling and linguistic research.

This paper provides an overview of Contemporary Arabic Lexicon (Mu’jam Arriyadh). It is a contemporary and inclusive Arabic dictionary that has been specifically developed to cater to the needs of both native and non-native Arabic speakers. The corpus utilized in this study is derived from the Arabic Contemporary Corpus for Analysis (ACCA), which encompasses a vast collection of 450 million words of Modern Standard Arabic spanning the previous century. Significantly, the lexicon in question prioritizes lemma-based entries over root forms, hence enhancing its user-friendliness and adaptability across different contexts. The resource offers comprehensive linguistic data pertaining to a wide array of Arabic vocabulary, encompassing morphological, morph-syntactic, and semantic aspects. The Lexicon has been developed in accordance with the ISO 24613 standard, which improves its ability to be processed by machines and facilitates the utilization of natural language processing systems. The database encompasses a range of linguistic aspects, such as synonyms, antonyms, and root forms, offering a comprehensive compilation. Mu’jam Arriyadh is a contemporary Arabic lexicon that is designed to be accessible to users, compatible with machine processing, and highly beneficial for anyone studying the language, conducting research, and utilizing natural language processing technologies.

Arabic diacritics encode phonetic information essential for pronunciation, disambiguation, and downstream applications, yet most Arabic ASR systems generate undiacritized output. In this work, we study direct speech-to-diacritized-text recognition using a single-stage ASR pipeline that predicts diacritics jointly with Arabic letters, without text-based post-processing. We evaluate two Arabic-adapted ASR architectures—wav2vec 2.0 XLSR-53 and Whisper-base—under a unified experimental setup on the ClArTTS Classical Arabic dataset. Performance is assessed using surface and lexical WER/CER alongside diacritic error rate (DER) to disentangle base transcription accuracy from diacritic realization. Our results show that Arabic-adapted wav2vec 2.0 achieves substantially lower diacritic error rates than Whisper, indicating stronger exploitation of acoustic cues relevant to vowelization. We further analyze the effect of decoding strategy and provide a detailed breakdown of diacritic errors, highlighting challenges associated with short vowels and morphosyntactic markers. These findings underscore the importance of model architecture and Arabic-specific adaptation for accurate diacritized Arabic ASR.

Automatic detection of toxic and offensive content in Arabic social media is a challenging task due to rich morphology, dialectal variation, and noisy writing styles. While transformer-based language models have achieved strong performance, they often produce uncertain predictions in borderline cases. This paper presents a hybrid framework for Arabic toxicity detection that combines a pretrained Arabic-specific transformer model with a confidence-aware rule-based mechanism. The proposed approach activates automatically induced lexical rules only when the model prediction falls within a predefined gray zone of uncertainty, preserving neural dominance while improving robustness and interpretability. Experiments conducted on a manually annotated dataset of 35,000 Arabic posts demonstrate that the hybrid approach achieves consistent improvements over the baseline model, particularly in reducing false negatives for toxic content. The results indicate that selective rule activation is an effective strategy for enhancing reliability in real-world Arabic social media moderation systems.

This research delves into the issue of hallucination detection in Large Language Models (LLMs) using Arabic language datasets. As LLMs are increasingly being used in various applications, the phenomenon of hallucination, which refers to generating factually inaccurate content despite grammatical coherence, poses significant challenges. We participate in the OSACT 2024 Shared-task (Detection of Hallucination in Arabic Factual Claims Generated by ChatGPT and GPT4). We explore various approaches for detecting and mitigating hallucination, using models such as GPT-4, Mistral, and Gemini within a novel experimental framework. Our research findings reveal that the effectiveness of these models in classifying claims into Fact-Claim, Fact-Improvement, and Non-Fact categories varies greatly, underscoring the complexities of addressing hallucination in morphologically rich languages. The study emphasizes the need for advanced modelling and training strategies to enhance the reliability and factual accuracy of LLM-generated content, laying the groundwork for future explorations in mitigating hallucination risks. In our experiments we achieved a 0.54 F1 in GPT-4 LLM.

Sarcasm detection and sentiment analysis are important tasks in Natural Language Understanding. Sarcasm is a type of expression where the sentiment polarity is flipped by an interfering factor. In this study, we exploited this relationship to enhance both tasks by proposing a multi-task learning approach using a combination of static and contextualised embeddings. Our proposed system achieved the best result in the sarcasm detection subtask.

Twitter and other social media platforms offer users the chance to share their ideas via short posts. While the easy exchange of ideas has value, these microblogs can be leveraged by people who want to share hatred. and such individuals can share negative views about an individual, race, or group with millions of people at the click of a button. There is thus an urgent need to establish a method that can automatically identify hate speech and offensive language. To contribute to this development, during the OSACT4 workshop, a shared task was undertaken to detect offensive language in Arabic. A key challenge was the uniqueness of the language used on social media, prompting the out-of-vocabulary (OOV) problem. In addition, the use of different dialects in Arabic exacerbates this problem. To deal with the issues associated with OOV, we generated a character-level embeddings model, which was trained on a massive data collected carefully. This level of embeddings can work effectively in resolving the problem of OOV words through its ability to learn the vectors of character n-grams or parts of words. The proposed systems were ranked 7th and 8th for Subtasks A and B, respectively.

Social media platforms such as Twitter offer people an opportunity to publish short posts in which they can share their opinions and perspectives. While these applications can be valuable, they can also be exploited to promote negative opinions, insults, and hatred against a person, race, or group. These opinions can be spread to millions of people at the click of a mouse. As such, there is a need to develop mechanisms by which offensive language can be automatically detected in social media channels and managed in a timely manner. To help achieve this goal, SemEval 2020 offered a shared task (OffensEval 2020) that involved the detection of offensive text in Arabic. We propose an ensemble approach that combines different levels of word embedding models and transfers learning from other sources of emotion-related tasks. The proposed system ranked 9th out of the 52 entries within the Arabic Offensive language identification subtask.