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 · pagefkxn0226_9.png

What is the title of the given bar graph?

OCR text of the page · 350 characters; the scanned image itself is not published
Average Age of Product Line
2004 - 2014
10
Revisions:
.........
April and
Revision:
9
November 2011
July 2014
Average Age
of Cards
Implants:
October and
December 2009
.....
W
. . ...
.......".
N
A
2004
2005
2006
2007
2008
2009
2010
2011
2012
2013
2014
ACCUMYN
DM
P-09870-00009
9
consulting
Source: https://www.industrydocuments.ucsf.edu/docs/fkxn0226
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
average age of product lineAverage Age of Product Line
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

Average Age of Product Line

27 characters6 tokens

Aux 2015

Wrong

ACCUMYN

7 characters4 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

OCR text:

Average Age of Product Line 2004 - 2014 10 Revisions: ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ... ...

151 characters64 tokens