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 · pagejxyn0081_1.png

Of all the groups, which group has the highest degree of self esteem?

OCR text of the page · 1,560 characters; the scanned image itself is not published
Segment Profiles
2. Play-Boys
Key Stats:
18% of the GenX population
Predominantly male (70%)
Youngest segment (average age of 21)
Least ethnically diverse (84% white)
Compared to other Xers, Play-Boys are:
Most Likely to Describe
Least Likely to Describe
Themselves as
Themselves as:
Easy going
" Practical
Open to new ideas
< Family-oriented
A pleasure seeker
Neighborly .
Adventurous
A smart shopper
Sneak Preview:
As the name implies, Play-Boys enjoy life. They have fun, live for today, and definitely
believe in the motto "pleasure before duty."
Of all the groups, Play-Boys possess the highest degree of self-esteem. While they
don't have a lot of money (yet), they place a great amount of emphasis on its
importance. Play-Boys are into material possessions are proud to it. As for working, a
job/career is definitely a means to an end for these guys - preferably, a weekend.
Play-Boys like to think of themselves as active and adventurous. They want more
excitement and sensation out of life, and like to imagine themselves doing things
others wouldn't dare do (like bungee jumping). Their vacations are on the active "field
and stream" side and their activities tend to be of a more strenuous nature (hiking,
camping, skiing)
While they don't like to shop, they certainly have a lot of stuff, and they like to be the
first to try new things (regardless of price). Play-Boys report the highest incidence of
ownership of VCR's, video games, CD players, personal computers and other
miscellaneous electronic devices.
H:CNXNIGENX\GXHILITE.DOC
11
51337 8099
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
Play-Boys
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

Play-Boys

9 characters5 tokens

Aux 2015

Exact match

Play-Boys

9 characters5 tokens

PiT-FT 2015

Wrong

<|<assistant|> <|<assistant|> <|<assistant|> <|<assistant|> <|<assistant|> <|<assistant|> <|<assistant|> <|<assistant|>

120 characters64 tokens

ChronoGPT 2015

Wrong

Play-Boys

Segment Profiles 2. Play-Boys Key Stats: 18% of the GenX population Youngest segment (average age of 21) Least ethnically diverse (84% white)

Question:

Of all the groups, which group has

202 characters64 tokens