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 · pagetnbf0227_1.png

Which is the State being surveyed?

OCR text of the page · 1,056 characters; the scanned image itself is not published
Hemoglobin data - Massachusetts
8% of the surveyed population had unsatisfactory hemoglobin levels
(ICNND guidelines) .
Males
Females
Age
Def.
Low
Total
Def.
Low
Total
Unsatisfactory
Unsatisfactory
0-5 yr
4.5
8 . 0
12 .5
1 . 5
6 - 12
7.0
0 .2
3 . 8
8.5
4.0
0 . 2
5.0
13-16
5 . 2
3 .6
12 . 7
16.3
0.0
3 .5
3 .5
17-59
1 . 2
10.0
11. 2
1 .1
6.0
7 .1
60+
0 . 7
14 . 3
15.0
0.5
4.7
5 . 2
National Nutrition Survey
Hematocrit data - Massachusetts
9.2% of the surveyed population had unsatisfactory hematocrit levels
(ICNND guidelines) .
Males
Females
Age
Def.
Low
Total
Def.
Low
Total
Unsatisfactory
Unsatisfactory
0-5 yr
4.4
4.0
8.4
6-12
0 . 5
1 . 5
2.0
0.0
3.9
3.9
0.0
5.4
13-16
5.4
1. 2
15.0
16.2
0.0
17-59
2.9
2 . S
0 . 7
10.0
10.7
0 .5
7.8
8 . 3
60+
0.0
20.0
20.0
0.5
3.8
4. 3
National Nutrition Survey
General Socio-economic data - Massachusetts
Total number of persons examined
4,568
Mean family size
5.31
Mean income: $ per annum
6, 500
Mean poverty index ratio
2.29
% of families below poverty index ratio of 1.00
19.5
rydocuments.ucsf.edu/docs/tnbf0227
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
Massachusetts
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

Massachusetts

13 characters3 tokens

Aux 2015

Wrong

We are given two sets of data from the National Nutrition Survey (NNS) for Massachusetts:

  • Hemoglobin data: 8% of the surveyed population had unsatisfactory hemoglobin levels
  • Hematocrit data: 9.2% of the surveyed population had unsatisfactory hematocrit levels
264 characters64 tokens

PiT-FT 2015

Wrong

<|assistant|<|assistant|<|assistant|<|assistant|<|assistant|<|assistant|<|assistant|<|assistant|<|assistant|<|assistant|<|assistant|<<assistant|<<assistant

155 characters64 tokens

ChronoGPT 2015

Wrong

Low Total Unsatisfactory Unsatisfactory 0-5 yr 4.4 4.0 8 . 4 6 . 4 8 . 4 6 . 4 0 . 5 4 . 5 8 . 5 0 . 5 13-16 5

110 characters64 tokens