Discover new work
Search beyond your library and decide which suggestions deserve a place in it.
Discover suggests research from sources outside your project library. Start from your project brief, a paper, folder, writing or a phrase you type. You can also explore work related to starred or read papers and research areas already present in the project.
A new project needs no papers to start: the project brief searches with the project’s name and description. Until the library has papers to link to, results are ranked on their match to the brief, their citations where the source reports them, and their recency. With the default filters, if nothing reaches the match threshold, Discover lists the closest matches and says so.
Results explain why they were suggested, including topic match, citation links, impact and recency when that evidence exists. You can shortlist, dismiss or add them. An external result is a candidate to review, not evidence that a claim is true or a research gap is novel.
Select +N on a candidate’s author line to see the full list. This does not select or add the candidate.
A scan uses your plan’s usage on Free and Pro, paid by the project owner. Saved results are free to read. A new seed scans automatically only when the payer has usage left. Weekly saved-search digests use the usage of the member who saved the search; when it is used up, the run is skipped with an explanation. Neither plan limits the number of scans or saved searches, subject to ordinary abuse-prevention limits. See Plans and credits.
Choose sources for a question
On the web, choose Describe it in words, enter a question and select the providers under Searching. Choose Search the literature to search those sources. For example, leave only Crossref checked to search Crossref alone.
Your checked sources stay selected when you submit a different question, including the first question from an empty search box. Rerunning the same question also uses the current selection. If you uncheck every source, no scan starts; choose at least one source before searching again. Switching to another seed loads that seed’s source settings. Weekly digests keep their source choices with their saved search.
Read the match evidence
Select a candidate to open Why this surfaced. Seed match explains its relationship to the seed you chose; the other factors show library links, citation impact and recency. A high overall score can include those other signals, so check the breakdown before deciding how relevant a paper is.
For a project-brief search, Distari also searches the project name and keywords to find candidates that a longer query may miss. Matching only a generic name word does not count as a perfect brief match. When the text matches only the name or keywords and none of the brief, the evidence says so. It lists the matched terms and identifies any score support from the source’s result order. When the score uses semantic similarity, Seed match explains that instead.
Saved scans keep their original scores. To refresh the ranking, choose Run the search again. Reloading the page only reopens the saved scan.
Understand the result tabs
On the web, Discover groups the results into four tabs:
- New contains candidates to review. Year, citation, match and open-access filters narrow this tab.
- In library contains matching papers your project already holds, including candidates you add from Discover.
- Shortlist contains candidates saved for later. Shortlisting uses no plan usage; choose Shortlist on a paper in New to save it here.
- Dismissed contains papers marked as not relevant, keeping them out of future scans for the project.
The narrowing filters apply only to New. A high match threshold does not hide papers from In library, Shortlist or Dismissed. The header Search field searches titles, authors and venues across every tab.
If a tab says No results match your search, try different text or clear Search to restore its results. You do not need another scan. An empty Shortlist or In library tab explains what belongs there; it does not mean the scan found nothing. In New, Loosen the filters restores the default narrowing choices while keeping your choice to hide papers already in the library.
Bring a result into the project
Open the result and inspect its metadata before adding it. Distari marks papers that are already in your library. Adding a new one uses the usual import, duplicate and processing rules, including permissions and plan usage.
The Add price is what processing one paper costs at your project’s depth, and Add all multiplies it by the papers in the Shortlist. A long paper can settle a little differently once its pages are counted. If the price changed since you opened Discover (for example, someone changed the project’s depth), Distari stops the add, adds nothing and shows the new price so you can try again. When the project’s automatic processing is off, Add only saves the record and costs nothing. Reference only and shortlisting also leave processing off.
Saved searches can send a weekly digest for a seed. Discover source coverage varies, and some publishers expose only abstracts or metadata. If a source is unavailable, try another seed or search directly in the library for work you already have.
For a structured analysis of papers already collected, compare them.