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 · pagejxkm0080_1.png

What is the ‘heading’?

OCR text of the page · 1,376 characters; the scanned image itself is not published
EXHIBIT I
BUSINESS ANALYSIS CONCLUSIONS
National Performance
. SALEM May YTD share (8.75) remains essentially flat
versus 1980 (8.73) but reflects a softening trend
due to :
Flatening trend on SALEM LIGHTS 85's and SALEM
ULTRA.
Increased rate of decline on FULL FLAVOR 100's.
Cumulative volume is up 4.78 versus YAG principally
due to the year-end 1980 load, but shipments evi-
dence a softening trend.
Geographic Performance
Share softness is not primarily a function of a
particular geographic area.
SALEM's greatest share loss did occur in the South
Atlantic, the area which represents our heartland
and the largest share increase in Newport's expan-
sion areas.
SALEM has above-average development in the South
Atlantic, North Central, and Mid Continent which
are good areas of menthol development.
Demographic Performance
18-24 year olds continue to represent SALEM's highest
index of switching gains and usage followed by the
25-34 age segment, and thus, both groups represent
the Brand's highest share of smokers.
In 1980, SALEM's ability to attract 18-24 year olds
has softened due to an erosion in FULL FLAVOR's
competitive appeal resulting in FULL FLAVOR. styles
share of smoker loss among this segment.
SALEM LIGHTS, especially the 85's style, continues
to demonstrate its competitive appeal to 18-24 year
olds, the group primarily responsible for Newport's
growth.
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
business analysis conclusionsBUSINESS ANALYSIS CONCLUSIONS
Models
4 of 4 columns · click a model to add or remove it

Ours

Partial · ANLS 0.74

EXHIBIT I BUSINESS ANALYSIS CONCLUSIONS

39 characters14 tokens

Aux 2015

Exact match

BUSINESS ANALYSIS CONCLUSIONS

29 characters8 tokens

PiT-FT 2015

Wrong

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

64 characters64 tokens

ChronoGPT 2015

Wrong

SALEM LIGHTS, especially the 85's style, continues to demonstrate its competitive appeal to 18-24 year olds, the group primarily responsible for Newport's growth.

162 characters34 tokens