The Best Tool for Literature Review? Research Rabbit vs Connected Papers!

Research Rabbit stands out as the best “free forever” tool for building a literature review workflow—but it comes with a steep learning curve and an interface that can feel overwhelming without a plan. Connected Papers is easier to navigate and can be especially useful for periodically checking prior work and derivative work, where its similarity-based network map tends to surface newer and older relevant papers more effectively.

Research Rabbit’s core appeal is cost and capability. It’s positioned as free for researchers indefinitely, and it can integrate with Zotero by importing an entire library or syncing collections. The tradeoff is usability: after uploading papers (the transcript describes a nine-paper collection), the interface can “vomit out” graphs, lists, and suggested authors in a way that makes it easy to lose track of what’s being explored. Users are advised to enter with a game plan—such as deciding whether to focus on “later work” after a set of seed papers, or using the tool to narrow the graph to a specific direction of inquiry.

Once users commit to that approach, Research Rabbit’s graph becomes powerful. The transcript describes how selecting suggested authors expands into long chains of related work, with nodes representing papers and authors and connections that are not always explicitly labeled as citation relationships. The layout can also feel counterintuitive: nodes can be dragged around, but the overall structure still behaves as a clustered system, and removing or restarting parts of the exploration can be necessary to change direction.

Connected Papers, by contrast, is presented as more intuitive from the start. Users input a keyword, paper title, DOI, or identifier, and the tool generates a network map. The key distinction is that this map is not a citation tree; it’s arranged by similarity. Node size reflects citation count, node color reflects publishing year, and clusters form where papers are strongly connected by similarity. The transcript highlights that the map’s positioning is not adjustable (no free X/Y axis movement), which can reduce misinterpretation and help users get a stable “big picture” view.

Where Connected Papers is most compelling is in prior and derivative work checks. The transcript claims Research Rabbit can do earlier/later work too, but Connected Papers more consistently surfaces older and newer papers—explicitly noting the appearance of a 2023 paper in Connected Papers while Research Rabbit’s later-work range in the example tops out before that. The recommendation is not to choose one tool exclusively: use Research Rabbit as a daily driver for deep exploration once the interface is learned, and rely on Connected Papers (often once or twice a month) to double-check blind spots in prior and derivative work.

Finally, the transcript adds a broader comparison: Litmaps is mentioned as a strong “daily driver” alternative because even its free tier supports limited inputs and allows changing axes, while still being easier to navigate. But for students who want free access and maximum exploratory power, Research Rabbit is the pick—provided they give it a few days to get past confusion and learn how to steer the graphs with a clear objective.