PRENA (Public Relations Engine Nusantara AI) is, to the best of our knowledge, the first dataset specifically designed for Indonesian digital public relations instruction tuning tasks.

Dataset Description

The dataset contains 748 instruction-response pairs in ChatML format, covering seven core PR task categories in Bahasa Indonesia. It combines real-world press releases scraped from Indonesian news sources with… See the full description on the dataset page: https://huggingface.co/datasets/ripkiiiii/prena.