Chinese Open Domain Knowledge Graph

Chinese open-domain knowledge graphs (CKGs) are large-scale structured repositories of information about Chinese entities and their relationships, aiming to facilitate various natural language processing tasks. Current research focuses on improving CKG quality and completeness, particularly through the development of multi-modal graphs incorporating images and the application of large language models (LLMs) for question answering and knowledge graph completion. These advancements leverage techniques like contrastive learning for data refinement and LLM-based frameworks for improved accuracy and efficiency in knowledge graph querying and inference, ultimately enhancing the utility of CKGs for applications such as question answering systems and information retrieval.