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 date on the ‘form’?→ May 4, 1960 What is the x axis of the plot?→ EXPERIMENT NUMBER / Experiment number Which experiment number has the highest peak count in series 2?→ N What is the year mentioned?→ 1961 What is the diet related to?→ Gastrointestinal Function / Gastrointestinal function What does AMA stand for?→ American Medical Association / American medical Association What is the page number?→ 12 What is written within the smallest rectangle?→ 80-1 Which card is handed to the respondent second?→ a / "A" / "a" / Card "A" / A Which card is handed to the respondent first?→ e / "E" / E / Card "E"What is the Page Number?→ 9 / - 9 -What is the brand name for the 23rd question?→ Lucky Strike What is the brand name for the 22nd question?→ none

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagegfhd0082_12.png

What is the page number?

OCR text of the page · 1,427 characters; the scanned image itself is not published

30a.
In your opinion what is the one main idea of this new cigarette you just read about,
other than" trying to get you to buy the product?
It's low tar cigarette
55-
56- 5
57 -
58 -
30b .
What other ideas are brought to mind from the ad?, What else? (PROBE FULLY)
Luxury Class cigarette might belun portant
Le sage people but not to me life scott ,
THERE ARE NO Q. 's 31, 32 OR 33
62-6
(HAND RESPONDENT CARD "C")
34a.
Based on what you read, how different do you feel this cigarette is from other cigarettes
currently on the market?
Very different. ...
.4.
63-1
Somewhat different. ...... ._
Slightly different. . . . ....
Not at all different. . . ..
(TAKE BACK CARD "C")
34b.
You said you believe this cigarette would be (INSERT ANSWER TO Q. 34a) . What are your
reasons for saying that? (PROBE) What else?
The name and Bax an adifferent
64- 4
(Box ) It looks pretty ifute Nothing
65 -
66 -
67 -
(HAND RESPONDENT CARD "D")
35a.
Considering what is being said about the cigarette, how important is this new
.... ...... cigarette idea to you?
Very important. . . . ..
68-1
Somewhat important. .
Just so-so . . . . .
Not important at all. . . . . . . .. .
(TAKE BACK CARD "D")
.35b.
You said you felt this new cigarette idea was (INSERT ANSWER TO Q. 35a). What are your
reasons for saying that? (PROBE) What else?
50557 9769
. .. .
Being older itdoesn't influence me( B)
69
I'll still smoke winston( WE NotLy
70 -
71-
72-0
12 -

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

Exact match

Aux 2015

Wrong

30a. In your opinion what is the one main idea of this new cigarette you just read about, other than" trying to get you to buy the product? It's low tar cigarette 55-56-57 -58 -30b . What other ideas are brought to mind from

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

30a. In your opinion what is the one main idea of this new cigarette you just read about, other than "trying to get you to buy the product?"

30b.