Connected Papers
Connected Papers builds a visual graph of papers similar to one you choose, using co-citation and bibliographic coupling over the Semantic Scholar corpus.
About this tool
Connected Papers takes a single origin paper and draws a graph of the few dozen most closely related works in its field. It is not a citation tree: papers are positioned by similarity, so two articles that never cite each other can still appear close together if they share many references and citing papers.
To build each graph it examines tens of thousands of candidate papers and measures similarity through co-citation and bibliographic coupling, then lays them out with a force-directed algorithm. Lists of prior works and derivative works help find foundational and follow-up research. Its data comes from the Semantic Scholar corpus, which covers hundreds of millions of papers across scientific fields.
What you can do
Similarity graph
Papers cluster by shared references and citations, revealing related work beyond direct citations.
Prior and derivative works
Lists highlight influential earlier papers and later works that build on the field.
Path highlighting
Selecting a paper shows its shortest similarity path back to the origin paper.
Broad coverage
Graphs draw on Semantic Scholar’s corpus, useful from machine learning to biology and philosophy.
Who is it for?
- A student gets a visual overview of a new field from one landmark paper.
- A researcher checks that no important paper is missing from a literature review.
- An applied scientist finds the foundational works behind a method they want to use.
Before you choose
The free tier allows five graphs per month, which suits occasional use but not intensive review work.
Check the plan terms: the Academic plan covers academic, non-profit and personal use, while for-profit work falls under the Business plan.
Frequently asked questions
Is Connected Papers free?
Yes. The free tier includes all features and five graphs per month. Paid Academic and Business plans remove the monthly limit.
Is Connected Papers a citation tree?
No. It arranges papers by similarity based on co-citation and bibliographic coupling, so related papers can appear together without citing each other.
Where does Connected Papers get its data?
It relies on Semantic Scholar’s open research corpus, released under the ODC-BY licence, which spans hundreds of millions of papers.
Check the official website for current features, pricing, and terms.
Similar tools
Semantic Scholar
ResearchSemantic Scholar is a free AI-powered search and discovery tool for scientific literature from Ai2, with an open API and datasets of papers and citations.
ResearchRabbit
ResearchResearchRabbit is a literature discovery tool that starts from seed papers and maps related work, authors and citations as an interactive graph.
SciSpace
ResearchSciSpace is an AI research assistant for academics, with literature reviews across 280M+ papers, chat with PDFs, writing help, citations and research agents.
Litmaps
ResearchLitmaps is a literature review tool that turns seed articles into interactive citation maps, recommends related papers and sends alerts on new research.
Undermind
ResearchUndermind is an AI research assistant for scientific literature that asks clarifying questions, reads hundreds of papers and follows citations to find what others miss.
Elicit
ResearchElicit is an AI research assistant that searches over 138 million academic papers and produces cited reports, tables and systematic reviews.