LegalViz is a novel dataset of legal relationship visualization task based on legal texts taken from EUR-LEX that store EU official legal documents.
LegalViz: Legal Text Visualization by Text To Diagram Generation.
This paper has accepted at NAACL2025.
Please visit paper from the link.
LegalViz dataset annotations are distributed under CC BY-SA 4.0.
When you create any derivations, e.g., datasets, papers, etc, from JDocQA… See the full description on the dataset page:
https://huggingface.co/datasets/mizuumi1/LegalViz.
