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 · pagepzbd0227_3.png

How long is each presentation on Tuesday 21,I session?

OCR text of the page · 845 characters; the scanned image itself is not published
- 2 -
II Session - P. M.
2:00 - 2:20 p. m.
Effects of Maternal Nutritional Status on:
(a) Foetal growth and development
(b) Neonatal morbidity and mortality
Dr. W. Tompkins
(c) Infant growth and development
d) Infant morbidity and mortality
Dr. S. Fomon
2:20 - 3:45 p.m.
Discussion
COFFEE BREAK
Second Day - Tuesday 21st
I Session - A. M.
9:00 - 10:40 a. m.
Reports of Current Studies of Maternal Nutrition
Status - 10 minutes each presentation
1) Central America
Dr. G. Arroyave
2) Chile
Dr. F. Monckeberg
3) Chile
Dr. C. Gomez Rog
4) Caribbean
Dr. D. B. Jelliffe
5) Mexico
Dr. J. Cravioto
6) Colombia
Dr. E. Cobo
7) U.S. A.
Dr. E. Gold
8) U. S. A.
Dr. R. Shank
9) Canada
Dr. G. Beaton
10:40 - 10:55 a. m.
COFFEE BREAK
10:55 - 11:35 a. m.
Factors affecting Maternal Nutritional Status:
(a) Dietary intake patterns - Miss E. Reh/Dr. G. Arroyave.
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
10 minutes
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

10 minutes

10 characters4 tokens

Aux 2015

Wrong
  • 2 -
5 characters5 tokens

PiT-FT 2015

Wrong

<|<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<

64 characters64 tokens

ChronoGPT 2015

Wrong
  • 2 - II Session - A. M.
  • 9 -

Discussion:

  • 2 - II Session - A. M.
  • 9 -

Question:

What is the average time for a presentation on Tuesday 21, I session?

Answer:

178 characters64 tokens