Preprint Bot

Developed by the Syracuse University Open Source Program Office

Never miss the paper that matters.

Preprint Bot reads every new arXiv paper each morning and surfaces the handful that semantically match your research instead of a list of keywords.

Profile Representation Learning cs.LG cs.AI
scanning arXiv · 1,284 today
How it works
01

Build a profile

Assemble your own corpus of research papers to match against.

02

Fetch nightly

New preprints are scanned the moment they appear on arXiv.

03

Match by meaning

Sentence-transformer embeddings rank papers by semantic similarity.

04

Morning digest

Top matches land in your inbox. Skim the day's reading in two minutes.

Features

Semantic search

Vector similarity finds papers you'd never have searched for.

Multiple profiles

Create a separate profile for every thread of your research.

Match your papers

Upload your own PDFs or search for papers on arXiv. Preprint Bot then reads them section by section for sharper matches.

Email digests

Daily, weekly, or monthly. Set your own frequency and score threshold.

Sign in with ORCID

Authenticate with your ORCID iD — no new password to remember.

Open source

GPL-3.0, built by Syracuse OSPO. Read the code, file an issue, run your own instance.

Start keeping up with scientific literature today.

Questions? Email us at preprintbot@syr.edu