You need to agree to share your contact information to access this dataset
This repository is publicly accessible, but you have to accept the conditions to access its files and content.
You agree to not use the dataset to conduct experiments that cause harm to human subjects.
Log in or Sign Up to review the conditions and access this dataset content.
Saraiki Assistant Chat SFT Dataset
Question-answering and conversational data for supervised fine-tuning of Saraiki language assistants.
About Saraiki
Saraiki (سرائیکی) is an Indo-Aryan language spoken primarily in Pakistan, particularly across southern Punjab and neighboring regions. It has a rich linguistic, literary, and cultural heritage, with its own vocabulary, grammatical structures, and regional varieties.
Despite its significant speaker community, Saraiki remains a low-resource language in Natural Language Processing (NLP), with relatively limited digital datasets, language models, and computational resources compared with widely supported languages.
About This Dataset
The Saraiki Assistant Chat SFT Dataset is a curated question-answering and conversational dataset containing user queries and assistant responses in strict Jataki Saraiki.
The dataset is designed specifically for Supervised Fine-Tuning (SFT) of language models to improve their ability to understand Saraiki user instructions, answer questions, maintain conversational context, and generate natural Saraiki responses.
It can be used to develop Saraiki conversational assistants, instruction-following language models, question-answering systems, and other Saraiki language applications.
Intended Uses
This dataset can support:
- Saraiki question answering
- Saraiki conversational assistants
- Chat-based language models
- Supervised Fine-Tuning (SFT)
- Instruction tuning
- Saraiki LLM training and fine-tuning
- Instruction following
- Conversational AI
- Dialogue generation
- Response generation
- Low-resource language model development
- Saraiki NLP research
- Saraiki language technology
- Academic and educational projects
Why This Dataset Matters
Instruction-following and conversational datasets are important for adapting language models to specific languages and improving their ability to respond naturally to user queries.
For low-resource languages such as Saraiki, there is a limited availability of high-quality question-answering, dialogue, and instruction-tuning resources.
This dataset aims to contribute to the development of Saraiki conversational AI and language models by providing supervised examples that can help models learn Saraiki question understanding, instruction following, and response generation.
Dataset Structure
The dataset contains conversational question-answering examples consisting of user inputs and corresponding assistant responses.
Each example may contain:
- User: A Saraiki question, instruction, or conversational message.
- Assistant: A corresponding Saraiki response.
The examples can be formatted for common supervised fine-tuning and chat-training workflows.
A typical conversational example follows the structure:
User: [Saraiki question or instruction]
Assistant: [Saraiki response]
Why This Dataset Matters
Question answering and instruction following are important capabilities for developing conversational and helpful language models. For low-resource languages such as Saraiki, the availability of high-quality question-answering and supervised fine-tuning datasets remains limited.
This dataset aims to contribute to the development of Saraiki NLP and language technologies by providing question-answering and conversational data for supervised fine-tuning, model training, evaluation, and downstream language-processing applications.
Dataset Structure
The dataset contains Saraiki user queries paired with corresponding assistant responses.
Each example consists of:
- User: A Saraiki question, instruction, or conversational message.
- Assistant: The corresponding Saraiki response.
The dataset can be used for supervised fine-tuning (SFT) of conversational language models, instruction-following models, and other NLP architectures.
Access Requirements
Access to this dataset is gated. Users are required to provide information about their intended use and agree to the applicable usage conditions before accessing the dataset.
By requesting access, users agree not to use the dataset to conduct experiments that cause harm to human subjects and agree to use the dataset for non-commercial purposes only.
Citation
If you use this dataset in research, publications, models, or publicly released projects, please cite and acknowledge the dataset:
@misc{raza2026saraikiassistantsft,
title = {Saraiki Assistant SFT Dataset},
author = {Raza, Muhammad Farjad Ali},
year = {2026},
publisher = {Hugging Face},
howpublished = {\url{https://huggingface.co/datasets/themohal/saraiki-assistant-sft}}
}
Disclaimer
This dataset is provided for research and educational purposes. Users are responsible for ensuring that their use of the dataset complies with applicable laws, ethical requirements, and the dataset's license and access conditions.
- Downloads last month
- 359