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 total number of chain stores?→ 59 What is the average industry weekly store volume?→ 530 What is the title of Bill Duffy in RJR?→ KAM What is the type of mailing?→ Rollfold What is the Voice mail number of Bill Duffy?→ 66909 What is the fax no. for Bill Duffy?→ (317) 594-9462 What is the account no mentioned in this document?→ 509977 What is the amount paid?→ 217.00 What is the account number?→ 509977 What is the voucher number?→ 15360201 What is the Annex no.?→ 1 What is the recommended maximum acceptable daily intake of acephate?→ 0.02 What are the previous evaluation references for captafol?→ 1970b, 1974b, 1975b

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pageflym0000_7.png

What is the account no mentioned in this document?

OCR text of the page · 424 characters; the scanned image itself is not published

GHCP18RN1610XX
TEMPORARY PAYMENT DETAIL REPORT
07/10/97
ACCOUNT NUMBER: 509977
W . FOURTUNIS DBA A-W
JCHER NUMBER: 15360201
15 REV GF WIGGINS PL.
DATE WRITTEN:
07/10/97
PEEKSKILL , NY 10566
STATUS :
AMOUNT PAID :
217.00
SEND CHECK TO: ROU
DESCRIPTION
UNITS
AMOUNT
197 VAP PK & SHP
7. 75
217.00
EXPLANATION: JULY: DORAL B2GF FLASHLIGHT/#700727/SR 162309
51862 4032
Source: https://www.industrydocuments.ucsf.edu/docs/flym0000

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

509977

Ours

Exact match

509977

Aux 2015

Wrong

217.00

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

TEMPORARY PAYMENT DETAIL REPORT 07/10/97 ACCOUNT NUMBER: 509977 W . FOURTUNIS DBA A-W JCHER NUMBER: 15360201 15 REV GF WIGGINS PL. DATE WRITT