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 · pagehlxc0081_1.png

What time is the ‘supper for subjects’?

OCR text of the page · 438 characters; the scanned image itself is not published
30OR
Things to do Today
Pm Group - 3/28 1 4/5188
12:30
Pm
Lunch -
brush teeth .
pm
1:00
ETS chamber ( blood ,
2 100
Zero Time ( in chamber )
Pm
out of chamber (
blood
to front lobby .
4:15
135 min saliva
4:30
150 min
saliva
5:60
190 min saliva
5:15
Supper for subjects
pm
6:30
270 min saliva
pm
8:00
B40 min saliva
4 hour urine
subjects may leave . Jake
") cooker wl gloo ice
2 )
10-24 br urine container
3 ) diary
*b 271W@
fa@
51136 5931
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
5:155.15P.m
Models
4 of 4 columns · click a model to add or remove it

Ours

Partial · ANLS 0.57

pm 5:15

7 characters7 tokens

Aux 2015

Partial · ANLS 0.67

pm

2 characters2 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

30OR Pm Group - 3/28 1 4/5188 12:30 Pm Lunch - brush teeth .

30OR Pm Group - 3/28 1 4/5188 12:30 Pm Lunch - brush

115 characters64 tokens