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 · pagexqxm0227_1.png

What is the name of the ‘director’?

OCR text of the page · 798 characters; the scanned image itself is not published
Sample #
REQUEST FORM
METABOLIC ASSESSMENT LABORATORY
Date of Service
Div. or Clinic
Vanderbilt University Med. Ctr.
Nashville, Tennessee 37232
Pt .
Peggy R. Borum, Director
Sandra Bennett, Supervisor
Age
Sex
Room 547 R. Light Hall
Unit #
Telephone: 622-6415
Diagnosis
Sample
Send Report To
Reported By
Date
Date Collected
Time
Mo.
Day Yr.
AM
PM
Description of Test
Results
Normal Range
Transferrin
Retinol Binding Protein
Vitamin C
Urine Niacin
N methylnicotinamide
I -methyl-2 pyridine-5-carboxamide
Whole Blood Folate
Serum Folate
Vitamin E
Vitamin A
Vitamin D
Vitamin B12
Riboflavin - RBC GSH Reductase
Thiamine - RBC Transketolase
I certify that the above laboratory assays are requested for research purposes.
Signature of requesting Physician Telephone
strydocuments. desi. CdurdotSixqxm0227
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
peggy r. borum
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

Peggy R. Borum

14 characters8 tokens

Aux 2015

Wrong

Director

8 characters2 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

ChronoGPT

Question:

What is the name of the ‘director’?

ChronoGPT

Question:

What is the name of the ‘director’?

ChronoGPT

135 characters64 tokens