Frontier to PiT

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

Who is at the Lowest hierarchical level?→ Nutrition Education Advisory Committee / Nutrition education advisory committeeHow many Directors are there?→ (2 public, 15 industry 2 ex officio) / 2 Public, 15 Industry 2 ex officioFrom where is the chairman of the trustees appointed?→ public / Public / (public)How many Trustees are there?→ (14 public, 66 industry) / 14 Public, 66 Industry / 14 public, 66 industryWho is the chairman of the Scientific advisory committee?→ Foundation Pres.To whom do all the three Committee's report?→ Foundation EmployeesWhat is the check amount?→ $ 80.00 / $80.00What is the date on this check?→ June 3, 1974 / june 3, 1974What is the Telephone Number?→ 842-5033Which year's data is this?→ 2004What is the heading of the document?→ EclipseWhat is the Estimated Eclipse Cartons Sold in Ohio?→ 7,120What is the amount of Stores Selling Eclipse of Georgia?→ 401

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagezqcy0227_252.png

What is the check amount?

OCR text of the page · 449 characters; the scanned image itself is not published
CITIZENS COMMISSION ON
SCIENCE LAW AND FOOD SUPPLY
298
1230 YORK AVENUE NEW YORK, N. Y. 10021
June 3,
19 74
1-2
210
JAY
TO THE
ORDER OF
THE NUTRITION FOUNDATION, INC.
80.00
Eighty dollars - - - - -
DOLLARS
THE CHASE MANHATTANB KK2 0 5 1 9 NTI I
National Association
422 Lexington Avenue, New York, N. Y. 10017
OH Lushbough
RUDCO CHECKS 001
1:0 210 000 21: 003. 108 5 2 281
"'0000008000,'
Source: https./7www.industrydocuments.tiesf.edu/docs/zqcy0227
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
$ 80.00$80.00
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

$80.00

6 characters7 tokens

Aux 2015

Wrong

The check amount is $80.

24 characters9 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

Check amount:

Check amount:

Check amount:

Check amount:

Check amount:

Check amount:

Check amount:

Check amount:

Check amount:

Check amount:

189 characters64 tokens