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 · pagelnmm0081_6.png

What is the adjective used here to describe Hillary Clinton?

OCR text of the page · 167 characters; the scanned image itself is not published
If Everyone Pays
SMOKERS
Why Do
Smokers Pay
MORE . .? ?
Hillary Rotten Clinton
EB...94
Says You Are
51359 5453
A Greater Hazard
To The Health Care Program
THAN........
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
rottenRotten
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

Rotten

6 characters3 tokens

Aux 2015

Wrong

smokers

7 characters3 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

Smokers Pay More than a hazard

Question:

What is the adjective used here to describe Hillary Clinton?

OCR text: If Everyone Pays SMOKERS

Smokers Pay

156 characters49 tokens