This dataset is designed for fine-tuning LLMs to extract time entities from the text, which is aimed to get the standard time string in json format.
It is divided into two parts:
general.json: Samples extracted from various news sources.
smartspeaker.json: Samples obtained from voice assistants.
First, extract the original time entity strings, which are then analyzed by a large model to… See the full description on the dataset page:
https://huggingface.co/datasets/dongrixinyu/TimeExtractor.
