SynthLaw-Dataset is a synthetic legal scenario dataset in complaint drafing.
The dataset consists of 4,532 samples. Note two keys in the synthetic process: 1) real-legal source configurations and supervision mechanisms in each interaction ensure that the generated data is aligned at the sentence level, closely… See the full description on the dataset page:
https://huggingface.co/datasets/ShengbinYue/SynthLaw.
