What is paperKB?

paperKB is a research knowledge base platform for organizing, searching, and querying biomedical literature. It helps researchers build curated collections of papers and interact with them using semantic search and AI-powered chat.

Key Features

Full-text ingestion — import papers from PubMed, DOIs, PDFs, or Zotero and automatically extract full text, citations, and structured observations.
Semantic search — find relevant passages across your corpus using natural-language queries, powered by vector embeddings and hybrid retrieval.
AI chat — ask questions about your literature and get answers grounded in the papers you've collected, with inline citations.
Citation graph — visualize how papers in your collection cite each other and discover related work through PageRank-based scoring.
Structured observations — automatically extracted gene-phenotype relationships and other facts, linked back to supporting evidence.
Literature map — an interactive visualization showing how papers in your corpus relate to each other by topic similarity.
Collaboration — share knowledge bases with groups, manage access levels, and work together on curated collections.

How It Works

Create a knowledge base — each KB is a focused collection of papers on a topic or project.
Import papers — add papers by PMID, DOI, URL, PDF upload, or Zotero sync. paperKB fetches full text and metadata automatically.
Search and explore — use semantic search to find relevant passages, browse the citation graph, or explore the literature map.
Chat — ask questions in natural language and get AI-generated answers grounded in your papers.

Who Is It For?

paperKB is designed for biomedical researchers, lab groups, and research teams who want to:

Build a searchable, structured archive of their reading
Quickly find relevant passages across dozens or hundreds of papers
Get AI-assisted answers that cite specific sources
Share curated literature collections with collaborators

Ready to get started? See the Getting Started guide.