Synthetic Persian QA Pair Classification (SynPerQAPC) is a Persian (Farsi) dataset created for the Pair Classification task, focusing on determining the correctness or relevance of an answer to a given question. It is part of the FaMTEB (Farsi Massive Text Embedding Benchmark). This dataset was synthetically generated using GPT-4o-mini to simulate real-world question–answering scenarios.
Language(s): Persian (Farsi)
Task(s): Pair Classification (QA Answer… See the full description on the dataset page:
https://huggingface.co/datasets/MCINext/synthetic-persian-qa-pair-classification.
