Urdu AI Dashboard

Monitoring AI enhancements in Urdu

Research Papers

Google Scholar papers focused on Urdu AI, LLMs, and datasets

2025 AI and LLM in Urdu

Between Myths and Metaphors: Rethinking LLMs for SRH in Conservative Contexts

Authors:

A. Humayun B. Zubair M. Mustafa- arXiv Preprint arXiv:2511.01907 2025

… scarcity of language resources forAIin low-resource settings, … To explore this challenge,we evaluateLLMperformance on a … Our data sample focuses on RomanUrdu(Urduwritten in …

2025 AI and LLM in Urdu

Cross-Lingual English–UrduSemantic Word Similarity Using Sentence Transformers

Authors:

I. Muneer A. Saeed…

… Research on semantic word similarity for South Asian languages, particularlyUrdu, isimmature. In recent years, transformer-based approaches have proved extremely successful for a …

2025 AI and LLM in Urdu

From Press to Pixels: EvolvingUrduText Recognition

Authors:

S. Arif S. Farid- arXiv Preprint arXiv:2505.13943 2025

…LLM-based text recognition. As part of this work, we introduce a new manually annotateddataset ofUrdu… the first super-resolution model tailored forUrdutext, as well as our fine-tuned …

2025 AI and LLM in Urdu

ChatGPT's Ability to Answer Cancer-Related Basic Questions inUrdu: A Comparative Study with English Responses

Authors:

W. A. Khan M. Soomro M. Afzal A. Zaki

… Chat Generative Pre-Trained Transformer (ChatGPT) is a large language model (LLM),introduced by OpenAIin November 2022, that has been trained on vast datasets covering a …

2025 AI and LLM in Urdu

Fine-Tuning Large Language Models with QLoRA for Offensive Language Detection in RomanUrdu-English Code-Mixed Text

Authors:

N. Hussain A. Qasim G. Mehak M. Zain…

… For low-resource and morphologically rich languages likeUrduand its Romanized forms, …preprocessing to model the task of detecting offensive RomanUrdu-English text. Whereas …

2025 AI and LLM in Urdu

Mitigating Social Bias in English andUrduLanguage Models Using PRM-Guided Candidate Selection and Sequential Refinement

Authors:

M. U. R. Khan

… scores forUrduacross all methods, highlighting structural inequities in multilingualLLMtraining;… the landscape of computational linguistics andartificialintelligence. They now underpin …

2025 AI and LLM in Urdu

UE-NER-2025: A GPT-based Approach to Multilingual Named Entity Recognition onUrduand English

Authors:

M. Ahmad H. Farid I. Ameer F. Ullah M. Muzamil…

… We employed a hybrid annotation strategy that combinesAI-assisted pre-labeling with …guages likeUrdu—each sample was independently reviewed and corrected by two nativeUrdu-…

2025 AI and LLM in Urdu

Stylometry-driven framework forUrduintrinsic plagiarism detection: a comprehensive analysis using machine learning, deep learning, and large language models

Authors:

M. F. Manzoor M. S. Farooq A. Abid- Neural Computing Applications 2025

… is based on theUrdulanguage, which …Urdu, their effectiveness in detecting intrinsicplagiarism presents a unique challenge. Despite the existence of someLLMmodels [8, 14] forUrdu, …

2025 AI and LLM in Urdu

UrduFactCheck: An Agentic Fact-Checking Framework forUrduwith Evidence Boosting and Benchmarking

Authors:

S. Ahmad H. Iqbal M. Ahsan N. Naeem…

… LLMs, and it serves as the first benchmark to measure the factuality ofLLManswers inUrdu.… forUrdu-to-English and Englishto-Urdutranslation. All translation is performed by anLLM…

2025 AI and LLM in Urdu

Leveraging LLMs for action item identification inUrdumeetings: Dataset creation and comparative analysis

Authors:

B. Sadia F. Adeeba S. Shams S. Hussain- Information Processing & … 2025

…Urdumeetings, has become crucial. To serve this purpose, this research presents the firstever dataset and guidelines for annotating action items in code-mixedUrdu-… recognizingUrdu…

2025 AI and LLM in Urdu

Paraphrase detection forUrdulanguage text using fine-tune BiLSTM framework

Authors:

M. A. Aslam K. Khan W. Khan S. U. Khan A. Albanyan…

… inUrdutext. An essential contribution of this work is the creation of a large-scaleUrduParaphrased … We suggest using a BiLSTM model forUrduparaphrase detection to close this gap. …

2025 AI and LLM in Urdu

Pipeline for Generating Large-Scale Synthetic RomanUrduQA Datasets: A Case Study on Saeed Ghani Herbal Products

Authors:

A. Rafi A. Miraj

…Urdusynthetic QA dataset assembly, driven by domain-focused enhancement and dualhuman-LLM… FAQs pages of the website usingAIsystems and presented through the Beautiful …