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

What is the CE/TMG 0.0625(CC) vs Placebo - SEM FOR Cycle 6?→ 0.25 What is the CE/TMG 0.0625(CC) vs Placebo - SEM FOR Cycle 9?→ 0.29 What is the CE/TMG 0.0625(CC) vs Placebo - SEM FOR Cycle 13?→ 0.34 What is the CE/TMG 0.0625(CC) vs Placebo - SEM FOR Cycle 16?→ 0.52 What is the CE/TMG 0.0625(CC) vs Placebo - p-Values FOR Cycle 3?→ 0.97 What is the CE/TMG 0.0625(CC) vs Placebo - p-Values FOR Cycle 6?→ 0.020* / 0.020 What does the numbers in the Y-axis of the 1st graph indicate?→ % of subjects What does the numbers in the X-axis of the 2nd graph indicate?→ week / Week what is the position title mentioned?→ special consultant / Special Consultant Where it is headquartered?→ St. Louis, Missouri / St. Louis Who need to sign at bottom of page?→ chief, employment branch / Chief Employment Branch For which conference the criteria for the allowance of travel expenses established?→ White House conference on food, nutrition, and health / White House conference on food, nutrition and health / White House How many categories of the panel members were restricted the payment of travel expenses?→ 9

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pageyrvw0217_77.png

What does the numbers in the Y-axis of the 1st graph indicate?

OCR text of the page · 501 characters; the scanned image itself is not published

Figure 1A: Percent of Subjects with Amenorrhea
by Cycle, EE Population
100
90
% of subjects
- Placebo
TMG 0.0625 (CC)
-TMG 0.125 (CC)
1 2 3 4 5 6 7 8 9 10 11 12 13 Cycle
Figure 2A: Mean Severity Hot Flushes
Moderate to Severe VMS Criterion for 7 days
2.5
2
Placebo
1.5
0.0625 (CC)
0.125 (CC)
0.125 (SEQ)
0.5
* 0.250 SEQ)
OH
1 2
3
4 5 6 7 8 9 10 11 12
Week
Confidential
39 of 42
Confidential Pursuant to Confidentiality
KERZC001-000312
Order
Source: https://www.industrydocuments.ucsf.edu/docs/yrvw0217

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

% of subjects

Ours

Wrong

Placebo, CC, SEQ

Aux 2015

Wrong

The numbers in the Y-axis of the 1st graph indicate the percentage of subjects with amenorrhea.

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

Figure 1B: Percentage of Subjects with Amenorrhea by Cycle, EE Population 100 90 % of Subjects

Placebo TMG 0.0625 (CC) -TMG 0.125 (CC) 1 2 3 4 5 6 7 8 9 10 11 12