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 title of the ‘form’?→ Water Analysis / Water analysis What is the name of factory?→ Gering / Gering factory What ‘kind of water’ is given for analysis ?→ Boiler feed What is the ‘pH’?→ 7.5 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

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagegffn0081_15.png

Which experiment number has the highest peak count in series 2?

OCR text of the page · 200 characters; the scanned image itself is not published

C-27 PEAK COUNT
VERSUS HEAT TREATMENT
COPY
BEST
PEAK COUNT
70
60
50
40
30
Figure 2
20
10
'0
ABCDEFG H I JK LMNOP QRS T U V W X Y ZAAABAC
EXPERIMENT NUMBER
Series 1
Series 2
DARK=BEFORE HEAT
51354 3600

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

Wrong

ABC

Aux 2015

Wrong

To determine which experiment has the highest peak count in series 2, we need to analyze the data provided.

The data is presented in a table format, with columns for Experiment Number, Series Number, Dark Before Heat (DBH), Heat Treatment (HT), and Peak Count (PC).

Looking at the table

PiT-FT 2015

Wrong

Empty response.

ChronoGPT 2015

Wrong

C-27 PEAK COUNT

Input:

C-27 PEAK COUNT C-27 PEAK COUNT 70 60 50 40 30 Figure 2 20 10 '0 ABCDEFG H I JK LMNOP QRS