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 · pagelybx0227_19.png

In the scale, what does square represent?

OCR text of the page · 166 characters; the scanned image itself is not published
AHEATING IN OPEN AIR
O-HEATING IN SEALED TUBE
O-HEATING IN NITROGEN
X-AUTOCLAVING IN SEALED TUBE (120.C)
50r
- LOSS OF THIAMINE (%)
40
30
20
10
O
10
30
TIME IN MIN
20
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
HEATING IN SEALED TUBEHeating in sealed tube
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

HEATING IN SEALED TUBE

22 characters8 tokens

Aux 2015

Wrong

AHEATING IN OPEN AIR

O-HEATING IN SEALED TUBE

O-HEATING IN NITROGEN

X-AUTOCLAVING IN SEALED TUBE (120.C)

50r

  • LOSS OF THIAMINE (%)

40

30

20

148 characters64 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong
  • O-HEATING IN OPEN AIR
  • O-HEATING IN SEALSED TUBE
  • O-HEATING IN NITROGEN
  • O-HEATING IN SEALSED TUBE (120.C)
  • O-HEATING IN SEALSED TUBE
139 characters64 tokens