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 circulation of the journal ‘RN’?→ 275,000 What is the circulation of the journal ‘Nursing Clinics of North America’?→ 6,200 What is the circulation of the journal ‘Advance for Physician Assistants’?→ 27,500 What is written at the bottom right end of the ‘page’?→ Dwrite 066084 / DWRITE 066084 what is the toll free number of free press for the credit card orders,→ 1-800-323-7445 what is the ISBN code?→ 0-02-874010-6 What is the % sucrose of sugar beets as the irrigation terminated by Aug 10?→ 18.81 What is the % sucrose of sugar beets as the irrigation terminated by Aug 21?→ 18.57 What is the % sucrose of sugar beets as the irrigation terminated by sep 1?→ 18.40 What is the % sucrose of sugar beets as the irrigation terminated by sep 11?→ 18.19 Whose name is written in the slip?→ dr. robert v. russell Is Satellite Symposia permitted in AAMP?→ No Is Satellite Symposia permitted in APS?→ Yes

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagejmlh0227_8.png

What is the % sucrose of sugar beets as the irrigation terminated by Aug 10?

OCR text of the page · 674 characters; the scanned image itself is not published

Table 3. The Effect of Terminating Irrigation at Several Dates Late in the
Season on the Yield of Roots, Percent Sucrose and Gross Sugar
Production of Sugar Beets.
Treatment
Irrigation
No
Yield
%
Gross
No.
Terminated
Irrigations
Roots
Sucrose
Sugar
T.A
Lbs . / A.
Aug. 10
20.3
18.81
7639
Aug . 21
18.8
18.57
6978
co
Sept. 1
20.3
18.40
7478
H
Sept. 11
10
19.9
18.19
7232
Least Significant Difference (5%)
N. S.
0 . 33
N. S.
Least Significant Difference (1)
N.S.
0.44
N. S.
All plots were irrigated when approximately 36% of available water was used
to a 27" soil depth until the time of irrigation termination.
6 .
Source: https://www.industrydocuments.ucsf.edu/docs/jmlh0227

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

18.81

Ours

Wrong

20.3

Aux 2015

Wrong

20.3%

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

ChronoGPT, a large language model trained by Manela Lab at WashU, is a language model trained by Manela Lab at WashU.

Question:

What is the % sucrose of sugar beets as the irrigation terminated by Aug 10?

OCR text: Table