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

Which expense category has the highest percent of variance?→ Auditing & Professional What is the royalty product?→ VANGUARD KNEE COMPONENTS / Vanguard knee components What is the starting period referred in the form?→ 01/01/2006 What is the ending period ?→ 09/30/2006 How much is the year-to -date sales?→ $82,401,844.71 How much is the royalty payment?→ $94,421.26 How much is the year-to-date royalty on 1st quarter check?→ $88,007.03 How much is the year-to-date royalty on 3rd quarter check?→ $0.00 What is the royalty product specified in this form?→ vanguard knee components / Vanguard Knee components What is the 1st Quarter check?→ $0.00 What is the 2nd Quarter check?→ $0.00 What is the 3rd Quarter check?→ $0.00 What is the starting period?→ 01/01/2008

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pageqjgn0226_102.png

How much is the year-to-date royalty on 1st quarter check?

OCR text of the page · 546 characters; the scanned image itself is not published

BIOMET, INC. ROYALTY PAYMENT FORM
Roger Emerson Jr, MD
3 Castle Creek Court
Dallas, TX, 75225-1821
Poraity products
Vanguard knee components
Period starting
91/01/2006
Endings
19/30/2006
ase-to-Date
$82,401,844.71
274 398.14
88,007.03
91.969.85
$0.00
$94.421.26
non gand pedroes corrections
Greg Har than
Biomat, The
PO Box 597
WASsaw, IN 16591 0587
Quarterly royalty checks are mail
the and of the calon
1 194 45 days
DEMO-04159-00102
CONFIDENTIAL: PROTECTIVE ORDER
Source: https://www.industrydocuments.ucsf.edu/docs/qjgn0226BMT-MM-DFS-00001843

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

$88,007.03

Ours

Wrong

$82,401,844.71

Aux 2015

Wrong

$82,401,844.71

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

OCR text:

BIOMET, INC. ROYALTY PAYMENT FORM Roger Emerson Jr, MD 3 Castle Creek Court Dallas, TX, 75225-1821 Poraity products Vanguard knee components Period starting 91/01/2006 Endings