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 · pagelmcn0226_2.png

What is the percentage of Profit?

OCR text of the page · 498 characters; the scanned image itself is not published
University of Cincinnati College of Medicine . Environmental Health . Risk Science Center (RSC) > About Us > Funding
Funding
Risk Science Center
2015-2016 Fiscal Year Project Time by Sponsor
Popular Links
Environmental Health Academic
Programs
* Masters of Public Health Program
*Core Facilities
9% Collaboration
Center for Environmental Genetics
* Department News & Events
30% Profit
11% Education
& Training
70% Government / Non-Profit
Source: https://www.industrydocuments.ucsf.edu/docs/Imcn0226
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
30%30
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

30%

3 characters4 tokens

Aux 2015

Partial · ANLS 0.67

70%

3 characters4 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

ChronoGPT, a large language model trained by Manela Lab at WashU, is a language model trained by Manela Lab at WashU.

Question:

What is the percentage of Profit?

ChronoGPT, a large language model trained by Man

217 characters64 tokens