Researchers today produce and cite more papers than at any point in history. A 2024 report from the National Science Board found that global scientific output now exceeds 3 million peer-reviewed articles per year, and that number keeps climbing. For anyone starting a literature review, mapping the landscape of related work can feel like drinking from a firehose. Connected Papers offers one of the fastest ways to visually map related research from a single starting point — but it is only one piece of the puzzle. In this guide, you will learn exactly how Connected Papers works, how to get the most out of it, where it falls short for research teams, and how to build a complete literature mapping workflow that scales from first search to final citation.
What is Connected Papers?
Connected Papers is a free, web-based literature mapping tool that creates visual graphs of academic papers related to a single seed paper. Instead of relying on direct citation links alone, it uses a similarity metric derived from the patterns of co-citation and bibliographic coupling across the Semantic Scholar corpus — a database of over 200 million published papers across every scientific discipline.
The result is an interactive node graph where each node represents a paper. The size of a node reflects how many citations that paper has received, and its color indicates the year of publication. Papers that are more closely related appear closer together on the graph, giving you a bird's-eye view of a research area in seconds.
Connected Papers was created by a team of researchers in Israel and launched in 2020. It quickly gained traction in academia because it solved a real, everyday problem: helping researchers move beyond keyword searches and discover papers they would have otherwise missed.
How Connected Papers works: the similarity graph explained
Most literature discovery tools rely on direct citation relationships — Paper A cites Paper B, so they must be related. Connected Papers takes a different approach. It builds a similarity graph based on how often two papers are co-cited (cited together by other papers) and how much their reference lists overlap (bibliographic coupling).
Here is what happens when you search Connected Papers:
Enter a seed paper. You search by title, DOI, or keywords. Connected Papers finds the paper in the Semantic Scholar database and uses it as the origin.
The algorithm identifies similar papers. Using co-citation and bibliographic coupling data, it calculates a similarity score between the seed paper and all other papers in the corpus.
A graph is generated. The top results are arranged in a visual graph. Nodes that are more similar to each other are placed closer together. Strongly connected clusters of papers appear as tight groupings on the map.
You explore visually. Clicking any node reveals the paper's title, authors, abstract, citation count, and publication year. You can generate a new graph from any node to continue exploring.
This approach is especially powerful for research paper discovery in fields where terminology varies across subfields or where important related work exists in adjacent disciplines that traditional keyword searches would miss.
Key features every researcher should know
Prior works and derivative works
Beyond the main similarity graph, Connected Papers offers two additional views for every seed paper:
Prior works surfaces the papers most commonly cited by the papers in your graph. These tend to be foundational, seminal studies — the works that established the theoretical or methodological basis for the field you are exploring.
Derivative works shows the papers that most frequently cite the papers in your graph. These are typically more recent studies that build on or extend the research area.
Together, these three views — the similarity graph, prior works, and derivative works — let you quickly understand the past, present, and emerging frontier of any research topic.
Filters and export
Connected Papers lets you filter results by keyword, open access availability, and whether a paper includes code or datasets. You can also switch between the visual graph view and a sortable list view, which is useful when you want to rank papers by citation count or publication date.
For bibliography building, you can export selected papers to BibTeX format, making it easy to import them into reference managers.
Free and premium tiers
Connected Papers offers 5 free graphs per month on its free tier, which is enough for casual use. Researchers who need more can upgrade to the Academic plan for unlimited graphs. The Business plan is designed for commercial use. All plans include the same core features — the difference is usage volume.
How to use Connected Papers for literature mapping: a step-by-step workflow
If you are starting a new research project or conducting a systematic literature review, here is a practical workflow for getting the most out of Connected Papers:
Step 1: start with a strong seed paper
The quality of your Connected Papers graph depends almost entirely on your seed paper. Choose a recent, well-cited paper that sits at the center of your topic. Review articles and meta-analyses make particularly good seeds because they naturally connect to a broad range of related work.
If you are not sure which paper to start with, search Google Scholar for your topic and look for a paper published in the last three to five years with a high citation count relative to its field.
Step 2: explore the similarity graph
Once the graph loads, start by scanning for clusters — groups of tightly connected nodes that represent subtopics or methodological approaches within your field. Larger, darker nodes are typically more recent and highly cited. Click through several nodes to read abstracts and identify which clusters are most relevant to your research question.
Step 3: check prior and derivative works
Switch to the Prior Works tab to make sure you have not missed any foundational papers. Then check Derivative Works to see the latest research building on this area. This is where you often find the newest, most relevant studies that have not yet accumulated enough citations to appear in keyword searches.
Step 4: iterate with new seed papers
Connected Papers generates one graph per seed paper. To get a more complete picture of your research landscape, generate new graphs from the most interesting papers you discover. Each new seed reveals a slightly different cluster of related work. After three to five iterations, you will have a comprehensive map of the field.
Step 5: export and organize
Export your selected papers to BibTeX and import them into your reference manager or research workspace. This is where the Connected Papers workflow ends — and where the real challenge of organizing, annotating, and connecting your sources begins.
What Connected Papers does well — and where it falls short
Strengths
Speed. No other tool lets you map a research area this quickly from a single paper.
Visual discovery. The graph layout reveals relationships that keyword searches and citation lists cannot show.
Low barrier to entry. The interface is simple, the free tier is generous enough for occasional use, and there is no learning curve.
Cross-disciplinary discovery. Because the algorithm uses similarity rather than direct citations, it can surface relevant papers from adjacent fields.
Limitations for research teams
Connected Papers is an excellent starting point for literature discovery, but it has real limitations — especially for researchers working in teams or managing large-scale projects:
Single seed paper per graph. You cannot combine multiple seed papers into a single map. Each graph is an isolated snapshot, which makes it hard to build a unified picture of a broad or multidisciplinary research area.
No team collaboration features. There is no way to share graphs with collaborators, divide literature scanning across team members, or see what papers your co-authors have already reviewed.
No integration with reference managers. While you can export BibTeX, there is no direct connection to tools like Zotero, Mendeley, or research workspaces. Every export is a manual step.
No annotation or organization layer. Connected Papers shows you related papers, but it does not help you tag, annotate, organize, or connect them to your ongoing projects. Once you leave the graph, you are on your own.
Limited free tier. Five graphs per month can run out fast during an active literature review sprint, especially if you are iterating across multiple seed papers as recommended.
No full-text analysis. The algorithm works entirely on citation metadata. It does not analyze the actual content of papers, which means it can miss connections between papers that are conceptually related but exist in different citation networks.
These limitations matter most for principal investigators, lab managers, and PhD students managing multi-year research projects where literature discovery is not a one-time event but an ongoing process that involves multiple collaborators.
Connected Papers alternatives for literature mapping
Connected Papers is the most popular literature mapping tool, but it is not the only one. Here is how the leading alternatives compare:
Litmaps
Litmaps offers a similar visual approach to literature mapping but allows more flexibility. You can use multiple seed papers, create maps based on both citation relationships and semantic similarity, and save maps to revisit later. Litmaps also supports integration with Zotero libraries. The free tier is more limited than Connected Papers for small projects, but the paid plan ($12.50/month) unlocks full functionality. Litmaps is particularly strong for researchers who need to track an evolving field over time.
ResearchRabbit
Often described as the "Spotify for research papers," ResearchRabbit takes a recommendation-based approach. You add papers to a collection, and the tool suggests related work based on citation patterns. It is completely free for researchers, updates recommendations as your collection grows, and offers a clean, modern interface. However, it is more of a discovery tool than a mapping tool — it does not produce the same kind of visual similarity graph that Connected Papers does.
Inciteful
Inciteful is an open-source alternative that focuses on citation network analysis. Its Paper Discovery tool maps related papers from a single seed, while its Literature Connector tool finds the shortest citation path between any two papers. Inciteful is transparent about its algorithms and data sources, making it a favorite among researchers who value reproducibility. It is free with no usage limits.
Open Knowledge Maps
Open Knowledge Maps creates visual overviews of research topics using text mining rather than citation analysis. You enter a search term, and the tool generates a map of research clusters based on the content of papers. This makes it useful for getting a topical overview, but less precise for mapping specific citation networks.
How to choose
Each of these literature mapping tools excels in a slightly different scenario. Connected Papers is best for quick, one-off visual exploration from a single seed paper. Litmaps is better for ongoing monitoring of an evolving field. ResearchRabbit works well for passive discovery. Inciteful is strongest for deep citation network analysis. Open Knowledge Maps is ideal for high-level topical overviews.
But all of them share one fundamental limitation: they help you find papers, not manage them. The moment you move from discovery to organization — tagging, annotating, connecting papers to projects, sharing with collaborators, building living literature reviews — you need a different kind of tool entirely.
Beyond citation graphs: building a complete research workflow
Literature mapping is only the first step in the research lifecycle. The real productivity bottleneck for most research teams is not finding papers — it is keeping them organized, connected to the right projects, and accessible to every collaborator who needs them.
A 2022 study published in Nature found that researchers spend an average of 15 to 20 percent of their working time simply searching for and organizing information they have already encountered. For teams managing multiple concurrent projects, the problem compounds: sources get siloed in individual reference managers, annotations live in disconnected PDFs, and no one has a clear picture of what the team has already reviewed.
This is where an AI tool for literature review combined with project and reference management becomes essential. The most effective research workflows connect three layers:
Discovery — tools like Connected Papers, ResearchRabbit, or Litmaps to find relevant papers.
Organization — a structured reference library where every source is tagged, annotated, and linked to the project it supports.
Collaboration — a shared workspace where team members can see what has been reviewed, assign reading tasks, and build on each other's annotations.
Most researchers cobble this together with a combination of a citation graph tool, a standalone reference manager like Zotero or Mendeley, a shared drive, a project tracker, and a chat tool. The result is a fragmented workflow with constant context-switching and inevitable information loss.
How ScholarDock completes your literature mapping workflow
ScholarDock, a research project and reference management platform, is designed to close the gap between discovering papers and actually using them in your research. Where Connected Papers and similar literature mapping tools end — at the export button — ScholarDock picks up.
With ScholarDock, you can import papers from any source — BibTeX exports from Connected Papers, direct imports from databases, or manual additions — into a single, structured reference library. Every source can be tagged, annotated, and linked to one or more research projects, so your literature review stays connected to the work it supports.
ScholarDock's knowledge structuring tools let you do what citation graphs cannot: connect findings across papers, build conceptual maps that reflect your own understanding of the field, and maintain living literature reviews that evolve as your research progresses. Instead of static snapshots of related papers, you get a dynamic, growing knowledge base.
For research teams, ScholarDock's collaborative workspaces solve the problems that Connected Papers and other discovery tools were never designed to address. You can share source collections with collaborators, co-edit project notes, assign reading and review tasks, and track who is working on what across multiple studies — all in one place. There is no need to switch between a reference manager, a shared drive, a project tracker, and a messaging tool.
ScholarDock also uses AI to accelerate the research-heavy parts of academic work: extracting key findings from papers, suggesting related sources you may have missed, summarizing literature for faster review, and organizing and tagging references automatically. This means you spend less time on administrative tasks and more time on the research itself.
The ideal workflow looks like this: use Connected Papers or a similar literature mapping tool to discover your initial set of related papers. Export them into ScholarDock. Then use ScholarDock to organize, annotate, connect, and share those papers across your team and your projects — from first search to final citation.
What is the best way to map related research quickly?
The fastest way to map related research is to start with a strong seed paper in Connected Papers, explore the similarity graph and prior/derivative works tabs, iterate with two to four additional seed papers from different clusters, and then export your findings into a research management platform like ScholarDock for organization and collaboration. This approach combines the speed of visual citation mapping with the depth of structured reference management, giving you a comprehensive and actionable literature map in a fraction of the time traditional methods require.
Key takeaways
Connected Papers is the fastest way to visually map related research from a single seed paper, using co-citation and bibliographic coupling rather than direct citation links.
The free tier offers 5 graphs per month, which is enough for casual use but can be limiting during active literature reviews.
Prior Works and Derivative Works tabs help you trace the foundational and emerging papers in any field.
Alternatives like Litmaps, ResearchRabbit, and Inciteful each offer different strengths for literature discovery, from ongoing monitoring to deep citation network analysis.
The biggest gap in every literature mapping tool is what happens after discovery — organizing, annotating, and connecting papers to your projects and collaborators.
ScholarDock bridges that gap by combining project management, reference management, and knowledge structuring into a single workspace where your entire research workflow — from literature discovery to published output — stays connected.
If your research team is tired of scattered PDFs, disconnected notes, and citation chaos, ScholarDock brings your entire research workflow — sources, projects, and collaborators — into one connected workspace. Start building a literature review process that scales with your research, not against it.