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 · pagegmcd0003_1.png

Who should you call for pickup?

OCR text of the page · 826 characters; the scanned image itself is not published
PROMOTION MATERIAL APPROVALS
MPE
SV- 2
Relationship
Approval Deadline:
Brand(s) Salem, Vanlage, Low, now (501213, 501214, 501315, Sole
Item .
Description of Use-
To be packed with POP Premium offers
Originator.
B. D. Brown
Initials BOke Date 3/ 14/95 1
Legal
Approval No..
Comments_
Note changes for Tin's Ros Vantage and now
Approvals
Initials
Date
Comments
1. Promotion
A. Production
B. Manager
C. Group Manager/Director
3/4/95
D. Senior Director
2. Law
3. Brand Management
A. Brand Manage
B. Vice President Brand
Marketing
4. Tax
5, Senior Management
A. Senior Vice President
Marketing
B. Executive Vice President
Marketing
6. Sales
Buyer :
Print Shop
51142 9370
Call for Pick-up:.
Karen Hughes
4 Rey. Bidg.
6593
"Name )
(Extension)
RJRY USA Form 8202 - Rev. 05/98
Source: https://www.industrydocuRouNacsf.edu/docs/gmcd0003
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
Karen Hughes
Models
4 of 4 columns · click a model to add or remove it

Ours

Wrong

Print Shop 51142 9370 Call for Pick-up:. Karen Hughes 4 Rey. Bidg. 6593

71 characters35 tokens

Aux 2015

Wrong

To be packed with POP Premium offers

36 characters8 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

Applying for Pick-up

Applying for Pick-up

Applying for Pick-up

Applying for Pick-up

Applying for Pick-up

Applying for Pick-up

Applying for Pick-up

App

188 characters64 tokens