Citation maps and discovery

Keyword search finds papers that use the words you typed. Citation maps find papers that are connected to one you already trust, even when they use different words. That makes them useful when you enter a new subfield, when your keywords are not working, or when you want to check that you have not missed a foundational paper.

Most of these tools start from a seed paper. Connected Papers draws a graph where papers that share many references sit close together, with older work and recent follow-ups shown in different colors. Litmaps builds a map from one or more seed papers and can alert you when new papers connect to it. Inciteful builds a citation network and suggests related work based on how papers link to each other, and it can also find the papers that connect two separate topics.

scite takes a different angle. Instead of mapping which papers cite which, it looks at the sentences where a paper is cited and shows whether later work supports it, contrasts with it, or only mentions it. That helps when you want to know how a finding has held up, though the classification is automated and worth checking by reading the citing paper. Open Knowledge Maps turns a topic search into a visual map of research areas, built on open data, which helps when you want an overview before you pick seed papers.

Featured AI research tools: citation maps and discovery

SponsoredCitation maps and discovery This position is open Put your tool at the top of the citation maps and discovery page. Paid spots are always labeled Sponsored. From $99 a month. Get featured

What a featured listing costs ยท from $99 a month

4 AI research tools in citation maps and discovery

Connected Papers

Builds a visual graph of papers related to one starting paper, grouping similar work so you can spot clusters and older foundations.

  • Free plan or free tier

Litmaps

Citation maps that grow from seed papers, with monitoring that alerts you when new papers connect to your map.

  • Free plan or free tier

Open Knowledge Maps

Nonprofit, open source tool that turns a search into a visual map of research areas, built on BASE and PubMed.

  • Free plan or free tier
  • Open source

scite

by Research Solutions

Shows how a paper has been cited by later work, including whether citing statements support or contrast with its findings.

  • Shows sources for its answers

A few habits make these tools more reliable, and they take little time. Start from more than one seed paper, ideally from different groups or methods, so one starting point does not narrow your view. Remember that a central position on a graph reflects how often a paper is cited, which is not the same as being right, and widely criticized papers can sit in the middle of a map. Read the papers you find before you rely on them.

Citation maps also work well with a reference manager. Once you have a set of papers you trust, export them to Zotero, Mendeley or another manager so they stay with your notes and PDFs. Some tools can import a collection from your reference manager as the seed, which saves time when a review grows.

The tools below are listed in no particular order, each with a short description we wrote and a link to its official site.

Questions about citation maps and discovery

How many seed papers should I start with?

One works, but two or three from different groups or methods give a fuller picture. A single seed can lock you into one school of thought, since the map follows that paper's references and citations.

Does a large node mean a paper is important?

It means the paper is cited often. Citation counts reflect attention, which can include criticism. Read the paper and the work that cites it before you treat it as a key source.

Can citation maps replace a database search?

Not for formal reviews. Citation maps are good for finding related work you missed, and many reviewers use them as a supplementary step after the main database search, called citation chasing or snowballing.

How does scite decide whether a citation supports a paper?

It uses automated text classification on the sentence where the paper is cited. The labels are a useful guide, but they can be wrong, so open the citing paper and read the context before you report that a finding was supported or contradicted.

Related categories