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 type of report is this?→ transmission report / Transmission Report What is the 'number of pages including cover sheet'?→ 3 What is the total pages scanned?→ 3 What is the 'total pages confirmed'?→ 3 What does SF denote?→ store & forward / Store & forward What does MB denote?→ Send to Mailbox What is the page number?→ 13 the n values are number of subjects at?→ cycle 1 / Cycle 1 What is written within the square?→ PLACE STAMP HERE / Place stamp here Who is the letter addressed to ?→ UNICEF/ UN CENTER What is the zip code ?→ 63105 What is the state code for Missouri?→ MO What is the name of the city ?→ CLAYTON / Clayton

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pageyhdw0217_13.png

What is the page number?

OCR text of the page · 434 characters; the scanned image itself is not published

Figure 1 Mean daily number of hot flushes per cycle in efficacy evaluable population of
postmenopausal women treated with 1 mg 17B-estradiol (17B-E2) plus either 1.125 or 0.25
trimegestone (TMG), or 1 mg estradiol valerate (E2V)/1 mg norethisterone (NET). The n values are the
numbers of subjects at cycle 1
13
Confidential Pursuant to Confidentiality
OLIVS019-002748
Order
Source: https://www.industrydocuments.ucsf.edu/docs/yhdw0217

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

Ours

Wrong

Order

Aux 2015

Exact match

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

Figure 2 Mean daily number of hot flushes per cycle in efficacy evaluable population of postmenopausal women treated with 1 mg 17B-estradiol (17B-E2) plus either 1.125 or 0.25 trimegestone (TMG), or 1 mg estradiol val