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 · pagemkvw0217_6.png

What is the priority of the first Article?

OCR text of the page · 937 characters; the scanned image itself is not published
Trimegestone - Publication and Abstract Tracking Report
Wednesday, April 02, 2003
IV. Manuscripts - Premarin TMG
CONTA213-006267
Authors
Priority Article title
Journal
Timing
Status
Lobo R,
A1
300-US Lead paper
(3-28-03) Gary Grubb to contact authors for participation
Dorin M,
Covering vasomotor, VMI,
(4-2-03) DW preparing outline by mid April
Lewis V,
metabolic and basic safety
Simon J,
Grubb G
|Pickar J,
A2
300-US Endometrial, bleeding
3-28-03) Jim Pickar to contact authors for participation
Yeh I
and safety
(4-2-03) DW preparing outline by late April
Cunnane M,
Archer D
Lindsey R,
A2
300-US Bone mineral density-
3-28-03) Jim Pickar to contact authors for participation
Gallagher C,
Final data analysis
(4-2-03) DW preparing outline by late April
Harrison D
Zinaman M,
Pickar J
Page [Page] of [Pages]
Design Write, Inc.
Source: https://www.industrydocuments.ucsf.edu/docs/mkvw0217
Confidential Pursuant to Confidentiality
Order
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
a1A1
Models
4 of 4 columns · click a model to add or remove it

Ours

Wrong

300-US Lead paper

17 characters7 tokens

Aux 2015

Wrong

The priority of the first Article is A1.

40 characters11 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

OCTOBER, 03, 2003

OCTOBER, 03, 2003

OCTOBER, 03, 2003

OCTOBER, 03, 2003

OCTOBER, 03, 2003

OCTOBER, 03, 2003

OCTOBER

121 characters64 tokens