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 · pagejzbn0226_10.png

What is the first step in the flowchart?

OCR text of the page · 639 characters; the scanned image itself is not published
PRIORITIZATION PROCESS STEPS
Public Comment
Progression of Information Gathering
3. Reasonably available information
2. Voluntary Call-ins
3. Section So and Bd rules
Is information sufficient
4. Section 4 rules, orders, consent agreements
for priority decision
(if necessary, using hered approach!
is information sufficient
for scoping?
is additional information
Yes L
weneeded to designate?
cathar/ develop
information
Initiate Priority Designation
Propose Priority Designation
Finalize Priority Designation
High Priority
Low Priority
Risk Evaluation
Subject to Judicial Review
Source: https://www.industrydocuments.ucsf.edu/docs/jzbn0226
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
Prioritization Process Steps
Models
4 of 4 columns · click a model to add or remove it

Ours

Wrong

Public Comment

14 characters3 tokens

Aux 2015

Wrong

What is the first step in the flowchart?

40 characters11 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

Priority Designation

Section So and Bd rules

Section 4 rules, orders, consent agreements

Is information sufficient

Voluntary Call-ins

Procedure for priority decision

Section 4 rules, orders, consent agreements

Is information sufficient

Section 4 rules, orders

286 characters64 tokens