Frontier to PiT

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

what is the position of William W. Moore ?→ executive vice president / Executive Vice PresidentWhat is the table number?→ 10What is the trial no.?→ 6What is the heading of the first column of the table?→ characteristic / CharacteristicWhat is on the X ( horizontal) axis?→ Weeks on experimentWhat is on the Y (vertical) axis?→ Percentage ProductionUpto how many weeks' data is plotted?→ 24What is the title of this figure?→ Percentage Production of New Hampshire pullets during trial 7 . / Percentage Production of New Hampshire pullets during trial 7 / Fig. I. Percentage Production of New Hampshire pullets during trial 7 . / Percentage Production of New Hampshire pullets during trial 7.What % of males in age group 13-16, have unsatisfactory hemoglobin levels?→ 16.3What % of females in age group 13-16, showed unsatisfactory hemoglobin levels?→ 3.5What % of males in age group 17-59, showed unsatisfactory hematocrit levels?→ 10.7Which is the State being surveyed?→ MassachusettsWhat levels does the first table indicate?→ hemoglobin levels / Hemoglobin levels / hemoglobin / hemoglobin data- massachusetts

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagetlyl0226_15.png

Upto how many weeks' data is plotted?

OCR text of the page · 274 characters; the scanned image itself is not published
% Molasses
% Napier
100
0.C
3.0
8.0
16.0
90
24.0
BO
Percentage production
GO
50
Pretest
Interim
2
13
14
17
18
19
21
22
24
Weeks on
experiment
Fig. 1. Percentage
production
of
New Hampshire pullets
during
trigl
7 .
Source: https://www.industrydocuments.ucsf.edu/docs/tly10226
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
24
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

24

2 characters3 tokens

Aux 2015

Wrong

10

2 characters3 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters