If you're getting errors on PDF uploads, you may be experiencing one of Elicit's known limitations:
The paper contains many graphs/charts/images. Elicit sometimes has trouble extracting data from PDFs that contain a large number of non-text elements.
The PDF was made by scanning or imaging a printed document. If the PDF contains images of text rather than text, Elicit will not successfully read your file.
The paper contains unusual or odd formatting that makes it difficult for Elicit to parse.
Your file is too large. On Basic, the file limit is 20MB. On Plus or higher plans, the limit is 200MB.
Your filename may be too long.
There is much more text in the paper itself than in a typical academic paper, regardless of file size.
Try Premium PDF parsing
For the first three problems above, complex figures, awkward layouts, and heavy formulae, Premium PDF parsing often gets further than standard parsing. Open your Account Settings, scroll to the Advanced section, and switch it on, then try the paper again. It is available on every plan except Basic.
It reads every PDF with a heavier model while it is on, not only the one giving you trouble, so it uses more of your monthly usage across the board. Turn it back off once you are done. See Usage limits in Elicit.
Premium PDF parsing will not help with a scanned document that contains images of text rather than text, or with a file that exceeds your plan's size limit.
If your issue is not one of the above, please email us at [email protected] and attach the PDF you're having trouble with as well as a screenshot of the error you're receiving.