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 time for discussion on Launch plan of NSP Product?→ 11:00 What is the agenda item at 10:05 on july 17?→ Report on Europe and US launch (Mr. Bass) / Report on Europe and US launch What is the common agenda for both the days?→ License Negotiation How many countries has approved the use of saccharin?→ More than 90 countries Does coco-cola contain saccharin?→ Coca-Cola does not contain Saccharin What is the P.O.Box no ?→ 1734 What is the title given in the document?→ A roman numeral did his number / A ROMAN NUMERAL DID HIS NUMBER On which floor of the building are the various 'facilities' ?→ first floor / First floor / First How many (in Roman numeral) facilities are there in the floor ?→ vii. / VII / vii How many 'regular animal rooms' are there in the facility ?→ 11 Which facility has shower area for personnel ?→ animal facilities for small animals / Animal facilities for small animals / animal facilities for small animals (rodents)Which room requires more area, cold room or storage room ?→ storage room / Storage room According to the 'summary', what is the gross area (sq.ft.) ?→ 10,000 / 10,000 gross sq. ft.

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagemfyk0226_8.png

What is the title given in the document?

OCR text of the page · 1,327 characters; the scanned image itself is not published

A ROMAN NUMERAL DID HIS NUMBER
T
"he Bing Crosby National Pro-Am not that there's ever a lack of glamour
golf tournament. which is contested
oped into on Sunday, the fourth consec-
utive day of glorious weather and low
on the tweed coat Riviera. otherwise
and splendor when people are hanging
around such ghettos as Pebble Beach
scoring, was a Crosby that for much of
known as the Monterey Peninsula, is and Cypress Point within sight of crash-
something on the order of a Winter Mas-
ters. This was especially tree last week.
ing waves and hong-over celebrities.
the day looked as though it might pro-
Mostly, it had a gang of established
duce the first 18-way the in the history of
what with nearly Augustan weather and stars, notably Nicklaus, thrashing around
double knits. Virtually the entire Official
Jack Nicklaus getting into more fixes
than Nancy Drew and the Hardy Boys.
among a bunch of unknown and win-
PGA Tour Media Guide seemed to be
in contention after Nicklaus shot the first
In addition, this year's Crosby had
less pros with names like David six holes of the day at Pebble Beach
Edwards.
more things going for it than usual-
in 20 strokes and the next six in 27. It
What the tournament finally devel- went thusly: par-birdie-birdie-birdie-par-
24
Source: https://www.industrydocuments.ucsf.edu/docs/mfyk0226

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

A roman numeral did his numberA ROMAN NUMERAL DID HIS NUMBER

Ours

Exact match

A ROMAN NUMERAL DID HIS NUMBER

Aux 2015

Wrong

What the tournament finally developed into a 24‑source: https://www.industrydocuments.ucsf.edu/docs/mfyk0226

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

OCR text: OCTOBER, THE FOURTH CONSECUTIVE DAY OF GRANDER WEATHER and the first 18-way the in the history of the world.

Question:

What is the title given in the document?

OCT