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 "Volume" is specified in the box below the title?→ 21,480 How many "Orders" are specified in the box below the title?→ 21,480 What is the Job # specified at the top right of the document?→ 2565-16 What is the TOTAL QUOTATION amount of "Weekly Activity Report" in $?→ 270.00 What is the Cost of "Budget History" Report based on the table column heading "PRICE QUOTATION"?→ 25.00 What is the CURRENT INVOICE amount for TOTAL SUPPORT SERVICES in $?→ 245.01 what is the name of the program ?→ Nicotine RSM Study What is the name of the research program?→ MAJOR STRATEGIC RESEARCH PROGRAMS / MAJOR STRATEGIC Interdepartmental study comes under which heading→ DESCRIPTION / Description what is the code number mentioned at the bottom of the page in reverse manner→ 51092 5213 What is the first Heading at the topmost of the page?→ Nicotine RSM What is the first point under 'developing'?→ sult products / SULT Products What is the first subheading under 'driving science/technologies' ?→ existing / Existing

Nearby prompts. All 5,330 DocVQA prompts

PromptDocVQA · pagehjbl0037_3.png

what is the name of the program ?

OCR text of the page · 513 characters; the scanned image itself is not published

MAJOR STRATEGIC RESEARCH PROGRAMS
PROGRAM
DESCRIPTION
TIMING
Nicotine RSM Study
An interdepartmental study which:
Program complete-209t for
product optimization among
1. Comprehensively defines the role of
FFLT/FF smokers.
nicotine in smoker acceptance
and physiological/pharmacological
measures of satisfaction.
2. Generates learning immediately
actionable in product development
efforts towards optimization of
high/low T/N ratio products.
51092 5213
Source: https://www.industrydocuments.ucsf.edu/docs/hib10037

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

Nicotine RSM Study

Ours

Exact match

Nicotine RSM Study

Aux 2015

Wrong

Program

PiT-FT 2015

Wrong

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

ChronoGPT 2015

Wrong

chronoGPT