Export Paper Reference Lists From Semantic Scholar Into a Google Sheet

The Scenario

Your systematic review team identified 30 key papers as the foundation of the evidence map. The next step is reconstructing the bibliography of each one — every work it cites — so you can identify which sources appear across multiple papers and build the shared-citation graph. The team has the 30 paper IDs in a sheet. Nobody has a clean way to get from there to a flat reference table.

The bad version:

Open the Semantic Scholar detail page for paper 1, scroll to the References section, manually note or copy each cited work, switch back to the sheet, paste title and year for each.
Paper 1 cites 47 references. Paper 3 cites 112. The reference list doesn't export cleanly from the web interface — you're reading it off the screen.
By paper 5 you've been at this for three hours and the team is waiting on you to start the overlap analysis.

The reference list is the input for the actual work. Spending hours reconstructing it manually is time that belongs to the analysis, not the data collection.

The Easy Way: One Prompt in SheetXAI

SheetXAI is an AI agent that lives inside your Google Sheet. It reads the paper IDs in column A, calls Semantic Scholar's references endpoint for each one, and writes every cited work as its own row — with a column tying it back to the source paper.

Here is the prompt for this task:

For each paper ID in column A, retrieve its full reference list from Semantic Scholar and write each cited paper as its own row with title, year, and first author into a sheet called References

What You Get

A References sheet with one row per cited work across all 30 papers.
Columns: Source Paper ID (the paper from column A), Title, Year, First Author.
Works cited by multiple papers appear as multiple rows — one per source paper — so your frequency analysis runs cleanly.
Papers whose reference list returned no results from Semantic Scholar are noted in the Source Paper ID column rather than silently omitted.

What If the Data Is Not Quite Ready

Column A contains DOIs instead of Semantic Scholar paper IDs

For each DOI in column A, resolve it to a Semantic Scholar paper ID, then retrieve the full reference list and write each cited paper as its own row with title, year, and first author into the References sheet

You only want references to papers published after a certain year

For each paper ID in column A, fetch its reference list from Semantic Scholar, filter to cited works published after 2015, and write title, year, first author, and source paper ID into the References sheet

You need to flag references that appear in more than three source papers

After pulling all reference lists into the References sheet for each paper ID in column A, add a Count column showing how many of the 30 source papers cite each reference — and sort by Count descending

Resolve IDs, pull references, filter, and flag shared citations in one prompt

For each paper ID in column A, fetch the full reference list from Semantic Scholar, filter to works with at least 10 citations, write each to the References sheet with source paper ID, title, year, and citation count, and mark any reference that appears under more than 5 source papers as Core Citation in a Tags column

Try It

Get the 7-day free trial of SheetXAI and open any sheet with a column of paper IDs from your systematic review. Ask SheetXAI to expand the reference lists into a flat table — and start the overlap analysis the same afternoon instead of next week.