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

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagelpjm0223_56.png

What is QHS?

OCR text of the page · 729 characters; the scanned image itself is not published
Pfizer Pharmaceuticals Group
: .
. . . . "
PPI
Present Pain Intensity
.. .
OHS
Taken Every Nigh
PM
Taken in the Afternoon
PO
Taken by mouth
SAEs
Serious Adverse Events
SAP
Statistical Analysis Plan
SF-MPQ
Short Form McGill Pain Questionnaire
TCAs
Tricyclic Antidepressants
TOPS
Treatment Outcomes in Pain Survey
TM
Three Times Daily
VAS
Visual Analogue Scale
V/F
Apparent distribution volume following oral administration
WSS
Worldwide Safety Standards
.. . . .
. .
. ...
. .. .- .
-.
-. . .f
. . . .. . . . note
". . -
. . . . . .
. .. ..
.. . . ..
. . . .
Protocol ID A945-1008
38
Working Protocol inc. Amendment #2: August 7, 2002
CONFIDENTIAL
Pfizer_MPatel_0203747
Source: https:/AAAA.industrydocuments.ucsf.edu/docs/lpjm0223
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
TAKEN EVERY NIGHTTaken every night
Models
4 of 4 columns · click a model to add or remove it

Ours

Partial · ANLS 0.94

Taken Every Nigh

16 characters5 tokens

Aux 2015

Wrong

Pfizer Pharmaceuticals Group

: . . . . . " PPI Present Pain Intensity .. . OHS Taken Every Nigh PM Taken in the Afternoon PO Taken by mouth SAES Serious Adverse Events SAP Statistical Analysis Plan

199 characters64 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

Pfizer_MPatel_0203747

Question:

What is the most common type of pain?

OCR text: Pfizer_MPatel_0203747

Pfizer_MPatel_0203747

Question:

146 characters64 tokens