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 · pagesnyc0227_109.png

What is footnote 2?

OCR text of the page · 638 characters; the scanned image itself is not published
TABLE VII-1.
THE ORAL HEALTH STATUS OF MALE MILITARY DEPENDENTS BY AGE, FEDERATION OF MALAYA,
1962
Age
PT2/
Attrition
Recession
Debris
Calculus
OHT
(years)
No.
Mean
No.
Mean
No.
Mean
No.
Mean
No.
Mean
No.
Mean
No.
Mean
0-4
71
0. 000
71
0. 013
0. 000
70
0.0
71
1. 213
71
0. 014
71
1.227
5-9
29
0.20
29
0. 014
9
0. 000
29
0.0
29
1.293
29
0.017
29
1. 310
10-14
2
4.00
2
0. 000
2
0. 000
2
0.0
2
0. 100
2
1.000
2
2.000
Total and
Mean
102
0.14
102
0. 013
11
0. 000
102
0.0
102
1.231
102
0. 016
102
1.248
1/ Diseased, missing, filled.
2/ Periodontal Index.
3/ Oral Hygiene Index.
10 I
Source: https://www.industrydocuments.ucsf.edu/docs/snyc0227
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
PERIODONTAL INDEXPeriodontal index
Models
4 of 4 columns · click a model to add or remove it

Ours

Partial · ANLS 0.94

Periodontal Index.

18 characters6 tokens

Aux 2015

Wrong

The question asks for footnote 2. Looking at the OCR text, footnote 2 is the second footnote in the table. The text starts with "TABLE VII-1." and then lists the columns. The second footnote is labeled "Source: https://www.industrydocuments.ucsf.edu/docs/snyc022

262 characters64 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

Table VII-1. Age PT2/ Attrition Recession Debris Calculus OHT (years) No. Mean No. Mean No. Mean No. Mean No. Mean No.

118 characters64 tokens