Large language models (LLMs) produce fluent but factually incorrect outputs which become significant when such models are applied to local languages, such as Arabic, because of the language’s complex morphology, dialectal richness, and the insufficient material available for training and evaluation. The competition HalluScoring-2026 Track 1 aims to address these problems and consists of two tasks that demand hallucination detection in Arabic question-answering settings. To solve this issue, we transfer the knowledge of a powerful reference-based teacher that is a stacked ensemble of five fine-tuned encoder classifiers with dynamic lexical, embedding, and entailment characteristics onto a simpler reference-based student model.

This half-day tutorial provides a comprehensive overview of Legal Natural Language Processing (NLP) with LLM for participants with a basic understanding of Computational Linguistics or NLP concepts. We introduce how NLP can help analyze and manage legal text by covering five key topics: legal text analysis with LLM insights, legal text retrieval, rhetorical role identification, legal text summarization, and addressing bias and hallucination in legal tasks. Our goals are to explain why these tasks matter for researchers in the legal domain, describe the challenges and open problems, and outline current solutions. This proposed tutorial blends lectures, live examples, and Q&A to help researchers and students see how language technology and LLMs can make legal information more understandable and efficient.

One of the key challenges of deploying Large Language Models (LLMs) in multilingual scenarios is maintaining output quality across two conditions: factual correctness and linguistic fluency. LLMs are liable to produce text with factual hallucinations, solid-sounding but false information, and fluency errors that take the form of grammatical mistakes, repetition, or unnatural speech patterns. In this paper, we address a two-framework solution for the end-to-end quality evaluation of LLM-generated text in low-resource languages.(1) For hallucination detection, we introduce a retrieval-augmented classification model that utilizes hybrid document retrieval, along with gradient boosting.(2) For fluency detection, we introduce a deep learning model that combines engineered statistical features with pre-trained semantic embeddings using an attention-based mechanism.