A translated dataset from SciFact verifies scientific claims using evidence from the research literature containing scientific paper abstracts. The process of creating the VN-MTEB (Vietnamese Massive Text Embedding Benchmark) from English samples involves a new automated system: - The system uses large language models (LLMs), specifically Coherence's Aya model, for translation. - Applies advanced embedding models to filter… See the full description on the dataset page:
https://huggingface.co/datasets/GreenNode/scifact-vn.
