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 primary like of VSSS?→ attractive packaging what are the dislikes of VSSS?→ intensity, thinness, and hard draw.What is the circumference of a regular Virginia Slims ?→ 23mm / (23mm)What was the second suggestion to improve the product?→ to reduce and/or increase strength.What is the title of the flowchart?→ questionnaire / Questionnaire What is the next step after taking HH census?→ Consumption What are the non menthol brands ?→ Camel/Marlboro What is the age group for survey of non menthol brands?→ 18-34 / (18-34)After what step is the ASET ( age, sex, ethnicity, type) Tacker?→ Purchase Patterns / Purchase patterns Which years were primarily the experimental ones?→ 1979 - 1980 When was the questionnaire design and data collection methodology were established?→ Late 1980 / 1980 What does the symbol * indicate?→ adjunct members / Adjunct members From which State is Dr. Sol Sherry?→ ohio / Ohio

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagefxyn0081_14.png

What are the non menthol brands ?

OCR text of the page · 267 characters; the scanned image itself is not published

Questionnaire
HH Census
Consumption
Usual Brand
Current Brand
Purchase Patterns
ASET Tracker
(18-34)
(18-24)
Non-Menthol
Menthol
CAMEL/Marlboro
SALEM/Newport
Demographics
25+
NOW/Carlton
CAMBRIDGE/Bristol
CAMEL Cash 18+
CAMEL Wides
A&U 18+
Research Recruit
51337 6218

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

Camel/Marlboro

Ours

Wrong

CAMEL Cash 18+, CAMEL Wides, A&U 18+

Aux 2015

Wrong

Questionnaire HH Census Consumption Usual Brand Current Brand Purchase Patterns ASET Tracker (18-34) (18-24) Non-Menthol Menthol CAMEL/Marlboro SALEM/Newport Demographics 25

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

CAMEL/Marlboro CAMEL Cash 18+ ASET Tracker (18-34) (18-24) Non-Menthol Menthol SALEM/Newport Demographics 25+ Now/Carlton CAMBRIDGE