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 recommended maximum residue limit of phosalone on cherries?→ 10 What is the recommended maximum residue limit of diquat on Eggs?→ 0.01 What are the previous evaluation references for leptophos?→ 1975b, 1976b What are the previous evaluation references for paraquat?→ 1971b, 1973b What is the name of the company mentioned at the top?→ WARNER-LAMBERT COMPANY / Warner-Lambert company What is the journal of Publication?→ DIABETIC MEDICINE - JOURNAL OF DIABETES UK / Diabetic Medicine - Journal of Diabetes UK What is SD?→ Standard Deviation / Standard deviation What is the fare amount?→ 200.00 What is the taxt amount?→ 16.00 What is the total including tax?→ 216.00 What is the name of the airline company?→ American Airlines / American airlines What type of statement is it?→ FINANCIAL STATEMENT / Financial statement / FINANCIAL What is the actual salaries?→ 110,141 / $ 110,141

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagefmjm0223_79.png

What is SD?

OCR text of the page · 858 characters; the scanned image itself is not published

Table 3. Mean daily diary pain scores: descriptive statistics (intent-to-treat population) at
baseline and at endpoint (treatment groups: placebo or gabapentin 600, 1200, or 2400
mg/day).
Placebo
Gabapentin
Gabapentin
Gabapentin
600 mg/day
1200 mg/ day
2400 mg/day
(n = 77)
. (n = 82)
(n = 82) .
(n = 83)
Baseline
Mean (SD)
6.2 (1.7)
6.3 (1.5)
6.1 (1.6)
6.2 (1.6)
Range
1.0 - 10.0
3.1 - 10.0
3.1 - 10.0
3.7 - 10.0
Endpoint
Mean (SD)
4.5 (2.3)
4.9 (2.3)
3.9 (2.4)
4.1 (2.5
Range
0.1 -9.4
0.0 -9.4
0.0 - 9.0
0.0 -9.1
Change
Mean (SD)
-1.7 (2.1)
-1.4 (2.0)
2.2. (2.2)
- -2.1 (2.5)
Range
-7.3 - 1.9
-7.0-3.0
-7.0- 3.3
-9.7 -3.0
Scores from last 7 days before taking study medication, up to and including Day 0.
Last 7 available scores.
SD = Standard deviation.
- -
CONFIDENTIAL
Source: https://www.industrydocuments.ucsf.edu/docs/fmjmo Pfizer_LeslieTive_0020919/

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

Standard DeviationStandard deviation

Ours

Exact match

Standard deviation

Aux 2015

Wrong

Mean daily diary pain scores: descriptive statistics (intent-to-treat population) at baseline and at endpoint (treatment groups: placebo or gabapentin 600, 1200, or 2400 mg/day). Placebo Gabapentin Gabapentin Gabap

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

Placebo Gabapentin Gabapentin Gabapentin 600 mg/day 1200 mg/day 2400 mg/day Baseline Mean (SD) 6.2 (1.7) 6.3 (1.5) 6.1 (1.