Text-Guided Legal Knowledge Graph Reasoning

Recent years have witnessed the prosperity of legal artificial intelligence with development technologies. In this paper, we propose a novel application provision prediction (LPP), which aims to predict related provisions affairs. We formulate task as challenging knowledge graph completion problem, requires not only text understanding but also reasoning. To end, text-guided reasoning approach. collect amounts real-world data from Guangdong government service website and construct dataset called LegalLPP. Extensive experimental results on show that our approach achieves better performance compared baselines. The code are available in https://github.com/zjunlp/LegalPP for reproducibility.