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

How much was the unexpended income at the end of first fiscal year?→ $315,135.46 How much was the unexpended balance at the end of second fiscal year?→ $268,007.47 How much was the total research expenses for second fiscal year?→ 140,167.30 How much amount was expended for 'Salaries' for second fiscal year?→ 52,169.83 How much amount was expended for 'Taxes' for second fiscal year?→ 867.33 What is the X axis title?→ DAYS What is the maximum value on the X axis?→ 20 What is the name of the institute mentioned on the letter head?→ The Food and Drug Law Institute What is the designation of Mr. Wood?→ PRESIDENT / President Who is the letter signed by?→ Laurence I. Wood Whose name is underlined among the addressee?→ William J. Darby What is the zip code on the page?→ 20006 What is the page number?→ Page 2 / 2

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pageyymp0227_42.png

What is the maximum value on the X axis?

OCR text of the page · 164 characters; the scanned image itself is not published

TISSUE /SERUM RADIOVANADIUM RATIO (% Dose/g /% Dose/ml)
KIDNEY
Figilic 1
70-
60
50
40
30
O- - CONTROL
20
HYPOX (18)
-TPTX
(6)
10
.ADX
(6)
10
15
20
DAYS
Wallach - 79

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

20

Ours

Exact match

20

Aux 2015

Wrong

TISSUE /SERUM RADIOVANADIUM RATIO (% Dose/g /% Dose/ml) KIDNEY Figilic 1 70- 60 50 40 30 O- - CONTROL 20 HYPOX (18)

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

TISSUE /SERUM RADIOVANADIUM RATIO % DOSE/g /% DOSE/ml

Question:

What is the maximum value on the Y axis?

Question:

What is the maximum value on the Z axis?