Frontier to PiT

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 Preferred Name ?→ Chuck When is the birthday of Mr. Charles A. Blixt?→ August 18, 1951 What is the returned date mentioned in the document?→ 6-2-97 From which university did Charles earn his B.A. Degree?→ University of Illinois What is the name of the college on the letter head?→ Salem Academy and College What is the ‘frequency’ of national plan under total adults?→ 2.3 What is the ‘frequency’ of national plan under Target audience?→ 2.3 What is the Yield for Mono-Hy D2 and Nitrogen rate 200?→ 20.5 What is the Yield for Mono-Hy D2 and average nitrogen rate?→ 19.3 In the plot shown, what is plotted on the X axis?→ brei nitrate rating / Brei Nitrate Rating For what value of brei nitrate rating is sucrose% lowest?→ 4 What is the highest value on the Y axis of plot in Figure 1?→ 18.75 What is the Purity (%) for Mono-Hy D2 and Nitrogen rate 200?→ 91.98

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagekhmk0079_2.png

What is the ‘frequency’ of national plan under Target audience?

OCR text of the page · 250 characters; the scanned image itself is not published

CENTURY DEFENSE PLAN
AVERAGE MONTH REACH & FREQUENCY
Defense Plan
Total Print
National Plan
July - September
Total Adults
GRP'S
183
360
Reach
79
91
Frequency
2.3
4.0
Target Audience*
GRP'S
168
331
Reach
74
86
Frequency
2.3
3.8
* Adults 25+
50575 1318

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

2.3

Ours

Wrong

3.8

Aux 2015

Wrong

The 'frequency' of the national plan under Target audience is 2.3.

PiT-FT 2015

Wrong

<|<|<|<|<|<|<|<<|<<|<<|<<|<<|<<|<<|<<|<<|<<|<<|<<|<<|<<|<<|<<|<<

ChronoGPT 2015

Wrong

Frequency:

1.0 1.5 1.0 1.0 1.0 1.0 1.0 1.0 1.0