• Backward citations — the papers your seed article cites (its references).
• Forward citations — the papers that later cite your seed article.
• Similar works — papers that share references, keywords, or are frequently cited together, but may not directly cite or be cited by your seed paper .

Each paper appears as a node in an interactive graph, with lines (edges) connecting papers that cite one another. Nodes are color-coded — green for papers already in your collection, blue for newly discovered ones — so you can instantly see what you have and what you might want to add .

Key Visualization Capabilities

Interactive Citation Maps — You can click any node to see its title, authors, year, and abstract without leaving the map. Mousing over a node in the sidebar highlights the corresponding node in the graph, making it easy to match lists to visuals .

Forward and Backward Navigation — Instead of scrolling through a static reference list, you can follow citation trails in either direction. Click “earlier work” to see what influenced a paper, or “later work” to see how the field evolved afterward. Each click rebuilds the map around the new seed paper .

Similar-Work Recommendations — Beyond direct citations, ResearchRabbit uses the citation network to surface papers that co-occur in reference lists or share topical keywords. This is especially useful for finding work that is conceptually related but not directly connected by a citation .

Author and Topic Views — You can switch from a paper-centric map to a co-authorship network, showing how researchers in a field are connected. Topic clustering views reveal how ideas group together and shift over time .

Live Filtering — The visualization can be filtered by publication year, relevance score, or other criteria, helping you focus on the most influential or most recent work in a field .

How It Changes the Literature Review Process

The traditional literature review is linear: you search a keyword, scan results, pick a paper, skim its references, search again. ResearchRabbit makes the process exploratory and network-driven. Start with one strong seed paper, and the tool surfaces related work you might never have found through keyword search alone .

For example, a researcher studying AI in healthcare might start with a 2020 review article. ResearchRabbit would show: the 45 papers that review cited (backward), the 120 papers that have cited it since (forward), and another 30 topically similar papers that share key references. All of this appears in a single interactive graph — not a list of 195 disjointed results.

Limitations to Keep in Mind

ResearchRabbit is a discovery tool, not a citation database. It relies on publicly available metadata, so coverage may be weaker for older publications, non-English works, or papers behind strict paywalls. It also works best when you start with a well-chosen seed paper — a poor seed can lead to a noisy or irrelevant map .

• Backward citations — the papers your seed article cites (its references).
• Forward citations — the papers that later cite your seed article.
• Similar works — papers that share references, keywords, or are frequently cited together, but may not directly cite or be cited by your seed paper .

Each paper appears as a node in an interactive graph, with lines (edges) connecting papers that cite one another. Nodes are color-coded — green for papers already in your collection, blue for newly discovered ones — so you can instantly see what you have and what you might want to add .

Key Visualization Capabilities

Interactive Citation Maps — You can click any node to see its title, authors, year, and abstract without leaving the map. Mousing over a node in the sidebar highlights the corresponding node in the graph, making it easy to match lists to visuals .

Forward and Backward Navigation — Instead of scrolling through a static reference list, you can follow citation trails in either direction. Click “earlier work” to see what influenced a paper, or “later work” to see how the field evolved afterward. Each click rebuilds the map around the new seed paper .

Similar-Work Recommendations — Beyond direct citations, ResearchRabbit uses the citation network to surface papers that co-occur in reference lists or share topical keywords. This is especially useful for finding work that is conceptually related but not directly connected by a citation .

Author and Topic Views — You can switch from a paper-centric map to a co-authorship network, showing how researchers in a field are connected. Topic clustering views reveal how ideas group together and shift over time .

Live Filtering — The visualization can be filtered by publication year, relevance score, or other criteria, helping you focus on the most influential or most recent work in a field .

How It Changes the Literature Review Process

The traditional literature review is linear: you search a keyword, scan results, pick a paper, skim its references, search again. ResearchRabbit makes the process exploratory and network-driven. Start with one strong seed paper, and the tool surfaces related work you might never have found through keyword search alone .

For example, a researcher studying AI in healthcare might start with a 2020 review article. ResearchRabbit would show: the 45 papers that review cited (backward), the 120 papers that have cited it since (forward), and another 30 topically similar papers that share key references. All of this appears in a single interactive graph — not a list of 195 disjointed results.

Limitations to Keep in Mind

ResearchRabbit is a discovery tool, not a citation database. It relies on publicly available metadata, so coverage may be weaker for older publications, non-English works, or papers behind strict paywalls. It also works best when you start with a well-chosen seed paper — a poor seed can lead to a noisy or irrelevant map .