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 · pagermwn0226_52.png

Which version ( 'vs') shows a percentage growth of 14%?

OCR text of the page · 193 characters; the scanned image itself is not published
Total Platform Unit Growth
144%
44%
33%
14%
VS. 01
VS. 02
VS. 03
VS. 04
DM
P-09686-00052
PROTECTED DOCUMENT. DOCUMENT SUBJECTTO PROTECTVEVORDERtrydocuments.ucsf.edu/docs/rmwn0226
DEPUY023041972
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
vs. 04
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

VS. 04

6 characters6 tokens

Aux 2015

Partial · ANLS 0.83

VS. 01

6 characters6 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

OCR text:

Total Platform Unit Growth 14% 44% 33% 14% 14% 14% 14% 14% 14% 14% 14% 14% 14% 14% 14% 14%

104 characters58 tokens