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 · pageqtgl0226_2.png

What is the period of Statement of Expenditures?

OCR text of the page · 868 characters; the scanned image itself is not published
UNITED STATES CANE SUGAR REFINERS' ASSOCIATION
Statement of Expenditures
November 1, 1942 - October 31, 1943
compared with budget
and 1943-44 Proposed Budget
Proposed
Accounts
Budget
Expenditures
Budget
1942-1943
1943-1944
Salaries
$ 10,000.00
$12, 180.00
$ 9,000.00
Office Expenses
1,000.00
336.80
1,000.00
Printing
1, 000.00
2.53
500.00
Educational
2,000.00
727.47
2,000.00
Travel and Personal
Expense
2, 000.00
1, 316.23
3,000.00
Miscellaneous
3,000,00
150.00
1,000,00
Total
$ 19,000.00
$ 14, 713.03
$ 16, 500.00
Legal Expenses
Mr. Brobeck
$ 12,000.00
$ 12,000.00
$ 12,000.00
Expenses
6,000.00
8, 200.00
6,000.00
Sullivan & Cromwell
5,131.30
2,000.00
Miscellaneous
5,000.00
1,956.52
2, 500.00
Total
$ 23, 000.00
$ 27, 287.82
$ 22, 500.00
GRAND TOTAL
$ 42,000.00
$ 42, 000,85
$ 39,000.00
November 19, 1943
Source: https://www.industrydocuments.ucsf.edu/docs/qtg10226
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
November 1, 1942 - October 31, 1943
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

November 1, 1942 - October 31, 1943

35 characters21 tokens

Aux 2015

Wrong

The period of Statement of Expenditures is November 1, 1942 – October 31, 1943.

79 characters31 tokens

PiT-FT 2015

Wrong

<|assistant| <|assistant| <|assistant| <|assistant| <|assistant| <|assistant| <|assistant| <|assistant| <|assistant| <|assistant| <|assistant

141 characters64 tokens

ChronoGPT 2015

Wrong
Statement of Expenditures
Statement of Expenditures
Statement of Expenditures

The same 26-character fragment repeats 11 times until the token limit. Showing the first three.

311 characters59 tokens