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 · pagexmww0217_17.png

What is the Duration (cycles) for Study Number Phase III "300200"?

OCR text of the page · 1,399 characters; the scanned image itself is not published
Table 1 Summary of clinical phase II/Ill studies of Totelle Cycle and comparators
Study
Major endpoints
Treatment arms
Duration
Total
Number
Reference
number
(cycles)
number
evaluable
of
for bleeding
subjects
Phase II
2001
dose-ranging
2 mg E2 + TMG
3
38
14
(0. 10, 0.25 or
0.5 mg)
2002
dose-ranging
+ TMG
5/7
266
13
(0.05, 0. 10, 0.25 or
0.5 mg)
2003
extension of Study
2 mg E2 + 0.25 mg
5/7
134
200'
TMG
Phase III
3001
efficacy, safety
2 mg E2 + 0.25 mg
13/26
163
148
15
( endometrial safety
TMG
in extra 13 cycles
Totelle Cycle
165
153
with 236 subjects
Trisequens
159
145
3003
efficacy, safety
Totelle Cycle
13
320
296
16
Cycloprogynova /Cy
314
296
clocur
300400
hemostasis
Totelle Cycle
6/7
93
67
Femoston
91
72
300200
prevention of
Totelle Cycle
26
241
postmenopause
Femoston
242
227
230
osteoporosis,
efficacy, safety
300500
glucose and
Totelle Cycle
6/7
56
39
insulin metabolism
Placebo
56
n/a
2 mg E2 + 10 mg
58
36
MPA
1 .
All studies were carried out in a double-blind, randomized, multicenter parallel-
group design, with the exception of Study 2003 which was an open multicenter
study. All phase III studies were multinational
2 .
In the phase III studies, a total of 782 and 779 women for Totelle Cycle and all
comparators, respectively, were evaluable for bleeding
1 May, 2003
[Page]
Confidential Pursuant to Confidentiality
Produced for use in NJ HRT litigation.
DUBIDO02-007519
Order
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
26
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

26

2 characters3 tokens

Aux 2015

Wrong

The duration (cycles) for Study 2001 is 38, Study 2002 is 266, Study 2003 is 134, Study 2004 is 296, Study 2005 is 241

118 characters64 tokens

PiT-FT 2015

Wrong

<|<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<<

64 characters64 tokens

ChronoGPT 2015

Wrong

Totelle Cycle Phase II 2001 dose-ranging 2 mg E2 + TMG 3/7 136 14 (0. 10, 0. 25, or 0. 5 mg) 2002 dose-ranging

  • TMG 5/7 133 143
128 characters64 tokens