New LLM Knowledge Base GPTKB 2.0 Improves Entity Disambiguation

Researchers have developed GPTKB 2.0, a knowledge base derived from a large language model that addresses the issue of entity disambiguation. This new system constructs a knowledge base with 38.4 million triples and over 1.6 million entities, differentiating between homonyms and merging synonymous mentions. A web demo allows users to explore the knowledge base, audit the provenance of facts, and perform queries using SPARQL or natural language. AI

IMPACT Enhances the accuracy and auditability of LLM-generated knowledge bases, potentially improving downstream applications.

RANK_REASON The cluster describes a research paper detailing a new LLM-derived knowledge base. [lever_c_demoted from research: ic=1 ai=1.0]

Read on arXiv cs.AI →

• alphaXiv
• arXiv
• CatalyzeX
• DagsHub
• Gotit.pub
• GPTKB 2.0
• Hugging Face
• Influence Flower
• LLM
• ScienceCast
• SPARQL

• paper
• other

AI-generated summary · Google Gemini · from 1 sources. How we write summaries →