Elicit
An AI research assistant for literature reviews
A research assistant that searches over a hundred million academic papers, summarizes and extracts data from them, and automates parts of a systematic literature review with citation-backed answers.
Picked for Literature Review & Evidence Kit.
Elicit was built for the specific pain of a literature review: finding the relevant papers, reading enough of each to know if it qualifies, and pulling comparable data points out of studies that don’t report them in a consistent format. It searches its academic database, ranks results against a research question, and can extract structured data from a batch of papers into one table.
It suits academic researchers, grad students and anyone running a systematic review under time pressure, where the screening and extraction stages are usually the most tedious part of the process. The citation-backed answers matter here more than in a general-purpose chatbot — every extracted value points back to the sentence it came from, which is what makes the output usable in something as rigorous as a published review.
It’s not a substitute for reading, though. The underlying search only covers what’s indexed in its databases, so niche or very recent work can be missing entirely, and the extraction step — being LLM-driven — occasionally misreads a table or pulls the wrong number from a paper with several similar figures. Treat it as a way to move faster through the mechanical parts of a review, with a human still checking the parts that matter.
Features
Paper search
Full-text search across well over a hundred million academic papers and hundreds of thousands of clinical trials.
Data extraction tables
Pulls structured data from hundreds of papers at once into a table, with each cell backed by a supporting quote.
Systematic review support
Auto-generates screening criteria from a research question and ranks papers by likelihood of meeting them.
Chat with papers
Ask questions of a specific paper's full text and get a cited answer instead of a page reference to search manually.
Custom extraction columns
Define the field you need pulled from every paper, so a comparison table answers your question rather than a generic one.
Concept-based search
Matches on what a paper is about rather than on the words it happens to use, which is what surfaces the study filed under different terminology.
Use cases
- Scoping the existing literature on a research question before writing
- Screening hundreds of papers against inclusion criteria for a systematic review
- Extracting a comparable data point, such as sample size or outcome, across many studies at once
- Building a comparison table of methods and outcomes across a set of studies
- Finding relevant papers that use none of your search terms
- Checking whether a research question has already been answered before proposing it
Compare Elicit head to head
Side-by-side comparisons, on pricing, platforms and where each one wins.
Looking for something similar?
Other research tools worth comparing against Elicit.