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 · pagelslw0227_36.png

What is the no. of studies conducted in elderly females?

OCR text of the page · 557 characters; the scanned image itself is not published
TABLE VI
Whole Body Protein Synthesis in Human Beings
at Various Ages
Number
Body Protein
Group
Age.
of Studies
Synthesis
(g/kg body wt/day)
Neonate
1-46 days
10
17 . 4 + 7.9
Infant 1
10-20 months
6.9 + 1.1
Young Adult
Male
20-25 years
3.3 + 0.3
Female
18-23 years
2.6 + 0.2
Elderly
Male
65-72 years
3.2 + 0.7
Female
69-91 years
2.3 + 0.4
1 Taken from the data presented in a study by Picou and Taylor-
Roberts (1969) , other data were obtained in the authors'
laboratories, as described in the text.
SOL
https://www.industrydocuments.ucsf.edu/docs/Islw0227
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
5
Models
4 of 4 columns · click a model to add or remove it

Ours

Wrong

2.3 + 0.4

9 characters9 tokens

Aux 2015

Wrong

2

1 characters2 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

OCTOBER OCTOBER OCTOBER OCTOBER

31 characters19 tokens