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

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagetlyl0226_10.png

What is the trial no.?

OCR text of the page · 947 characters; the scanned image itself is not published
Table 10. Statistical analyses of data obtained in trial 6.
BETWEEN SUGAR LEVELS
BETWEEN LEVELS OF BAGASSE
CHARACTERISTIC
Significant
Degrees of
freedom
F value
Significant
Degrees of
F value
freedom
N Percentage hen-day ovulations . .....
3.005
No
2 and 4
0. 387
2 and 4
No
2 and 4
Percentage hen-day production....
3.210
No
0.455
No
2 and 4
Percentage meat and blood spots detected by candling . . . . ....
1.862
No
2 and 12
0. 007
No
2 and 2
Percentage soft shells and broken eggs ..........
0. 397
No
0.549
No
2 and 12
Percentage double-yolk eggs. ......
2 and 12
3. 401
No
2 and 12
8. 423*+
Yes
2 and 12
Average increase in egg weight, gm. ......
0.440
No
0.447
No
2 and 12
Pounds of feed per dozen ovulations. .
15. 427"
Yes
2 and 12
2 and 12
1.090
No
2 and 12
Pounds of feed per dozen eggs. .. . . .
16.8709*
Yes
2 and 12
1. 581
No
2 and 12
.* P<0.01 percent due to chance alone.
Source: https://www.industrydocuments.ucsf.edu/docs/tly10226
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
6
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

6

1 characters2 tokens

Aux 2015

Wrong

2

1 characters2 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

Table 10. Statistical analyses of data obtained in trial 6. BETWEEN SUGAR LEVELS BETWEEN LEVELS OF BAGASSE CHARACTERISTIC Significant Degrees of freedom F value Significant Degrees of freedom BET

195 characters64 tokens