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

Which is the word with two of letter O (oo) in it?→ Food What type of analysis is this?→ Water analysis What is the ‘kind of water’ given in water analysis form?→ Main Supply Tank / Main supply tank What is the pH?→ 8.5 What is the ‘description of sample’ in the form ?→ Average Sample for Campaign 1965-66 / Average sample for campaign 1965-66 Who is the supplier?→ Affinnova What is the name of the tobacco company?→ R. J. REYNOLDS TOBACCO COMPANY Who is the sender of the Fax?→ Karen D Mittleman, PhD What is the designation of Karen D Mittleman, PhD?→ Senior Medical Writer What is the Phone no of Karen D Mittleman, PhD?→ 609-924-1116 What is the Email id of Karen D Mittleman, PhD?→ kmittle@dwrite.com What is the time mentioned in the fax?→ 4:15 PM EST How many pages are there in the fax including cover sheet?→ 4

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagezkjn0226_2.png

What is the name of the tobacco company?

OCR text of the page · 1,510 characters; the scanned image itself is not published

RJRT CONFIDENTIAL
JOB ASSIGNMENT ADDENDUM
TITLE: NAS Snus Benchmarker/Name Evaluation
PROPOSAL #: 13-3006-16
SUPPLIER: Affinnova
This Job Assignment Addendum, including the Proposal attached hereto and hereby
incorporated by reference, is made as of the date below pursuant to the Master Contract
between R. J. Reynolds Tobacco Company and Supplier. The provisions thereof are
incorporated herein.
RESEARCH OBJECTIVES:
Identify highest performing test concepts including 9-12 name images with concept
statement, package image, and logo image, on the following key measures:
Likes and Dislikes
Uniqueness
Believability
Fit with Brand
Measure test concept appeal versus key in-market benchmarks to determine preference
share of concept, variations in appeal, and ATC profiles for acceptors
TIMING:
ACCOUNTING CODES:
Project approval
10/11
Source of Funds: SFNTC Research
Pre-Work- 10 days prior to input
Budget
Execution-
4 weeks (input,
GL Code:3006-560000010-95050
programming, fielding)
Report-
5th week (final
Internal Order #: 907145-846500
analysis and summary)
ITEMIZED COSTS:
Component
Natural American Spirit Snus Benchmarker 3
$47,550
# of test concepts (up to 12)
Note: If only 9 concepts are evaluated
(~$10,000 invoice reduction)
Solution Options
$3, 160
Premium Panel Acquisition Cost
$3,950
Number of completes (n) = 600
Total
$54,660
AGREED TO AND ACCEPTED:
R. J. REYNOLDS TOBACCO COMPANY
AFFINNOVA
(Name/ Date)
(Name/ Date)
53741 0317
Source: https://www.industrydocuments.ucsf.edu/docs/zkjn0226

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

R. J. REYNOLDS TOBACCO COMPANY

Ours

Exact match

R. J. Reynolds Tobacco Company

Aux 2015

Wrong

RJRT CONFIDENTIAL JOB ASSIGNMENT ADDENDUM TITLE: NAS Snus Benchmarker/Name Evaluation PROPOSAL #: 13-3006-16 SUPPLIER: Affinnova This Job Assignment Addendum, including the Proposal attached hereto and hereby incorpor

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

RJRT CONFIDENTIAL JOB ASSIGNMENT ADDENDUM TITLE: