Output Explorer

Every prompt in the paper, and what each model wrote back.

Answer a question about a scanned page from its OCR text. Scored by exact match and ANLS against the accepted answers.

13 of 5,330 prompts

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pageysbw0217_3.png

Which is the source?

OCR text of the page · 421 characters; the scanned image itself is not published
3
Agenda for Menopausal Health Publication Management Meeting
April 2004. 10:30 AM-12:00 PM. CVCRE5101
Status
Meeting
Author
Title
Source
SUBMITTED
5" International
Pace
Effect of hormone replacement
GMA
Symposium on
therapy on quality of life in
Women's Health and
postmenopausal African American
Menopause
women
Florence, Italy
April 21-24, 2004
DESIGN134416
Source: https://www.industrydocuments.ucsf.edu/docs/ysbw0217
System prompt · identical for every setup
Answer the question using only the OCR text from a single document page. Return only the answer, with no explanation. Preserve the answer wording from the OCR text when possible.
Expected answer
GMA
Models
4 of 4 columns · click a model to add or remove it

Aux 2015

Wrong

Meeting

7 characters2 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

ChronoGPT ChronoGPT is a large language model trained by Manela Lab at WashU.

Question:

Which is the source of the following text?

ChronoGPT ChronoGPT is a large language model trained by

194 characters64 tokens