ZSS: Zotero Similarity Selection

ZSS is a research support tool that helps scholars efficiently filter relevant papers from large Zotero libraries by computing semantic similarity using state-of-the-art sentence embeddings. Given a CSV export of a Zotero library, the tool compares paper titles and abstracts against a user-defined description of the research scope and automatically selects the most relevant entries.

**Figure 1.** ZSS interface for uploading a Zotero CSV export and defining the research scope description used for semantic similarity comparison.

The filtered results can be exported as a CSV file and converted to BibTeX for re-importing into Zotero. The results page provides insights about the selection process, including a similarity distribution histogram to visualize how papers are distributed across similarity scores and where the threshold is.

**Figure 2.** Results dashboard showing corpus statistics, similarity threshold, and the distribution of similarity scores.