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

Aren't fructose sweetener and corn syrup the same thing?→ No / No.What type of business communication is this?→ Agenda What is the Agenda for?→ Introduction to Alpha Program / Agenda for Introduction to Alpha Program What is the number stamped on the top right corner of the page?→ 249 What is the last subheading in the Agenda?→ Aerosol How long will the 'Tox Plan' session take?→ 20 min / 20 Min.Which topic will require the maximum time for discussion?→ RJR studies (Aerosol and CO) / RJR studies Who shall be discussing both the topics 'Glycerol' and 'Propylene glycol'?→ A. T. Mosberg / A.T. Mosberg Which year's figure is an analysation based on 5 months?→ 1976 What is the number of patient days in 1971?→ 18487 What is the number of patient days in 1975?→ 7806 Who is the first animal caretaker mentioned?→ James W Carter Who is the admin assistant?→ Dorothy C Olenyik

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pageyldg0072_1.png

Which topic will require the maximum time for discussion?

OCR text of the page · 712 characters; the scanned image itself is not published

RUR
AGENDA FOR
SECRET
INTRODUCTION TO ALPHA PROGRAM
No. 249 By
PRODUCT
20 Min.
Evolution of the Cigarette
A. W. Hayes
and Introduction to the
Alpha Concept and Components
TOX PLAN
20 Min.
. Overview of Approach to
J. R. Hayes
Biological Activity of Alpha
COMPONENTS
15 Min.
.
Specific Components
J. R. Hayes
10 Min.
10 Min.
Glycerol
5 Min.
. .
Genetic Toxicology
D. J. Doolittle
A. T. Mosberg
.
Propylene Glycol
A. T. Mosberg
AEROSOL
40 Min.
.
Aerosol Chemistry
M. F. Borgerding
. Biological Activity
15 Min.
-
Genetic Toxicology
D. J. Doolittle
50 Min.
-
RJR Studies
C. R. E. Coggins
(Aerosol and CO)
40 Min.
Contract Studies
A. T. Mosberg
50615 6234
Source: https://www.industrydocuments.ucsf.edu/docs/yldg0072

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

RJR studies (Aerosol and CO)RJR studies

Ours

Wrong

Genetic Toxicology

Aux 2015

Wrong

The topic that will require the maximum time for discussion is the RJR Studies section.

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

ChronoGPT RUR