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 · pagenpbb0079_11.png

What is the date of preparation ?

OCR text of the page · 1,421 characters; the scanned image itself is not published
RJRT
REQUEST FOR CHANGE
F-1001
7/88
RFC
$80031
ORIGINATOR Name:
Date Prepared
Page
Date
Carolyn R. Carpenter
8-24-88
1_ of /
8:24 -68
Position
Telephone No.
Received By
Associate R&D Technologist
741-2483
Department Name and Location
Change Order Number
R & D New Product Technology - BGTC
Brand/Assembly/Material Affected
Menthol Product Delivery Specification
Type of Change
Addition
Design
Spec Change
Emergency
Deletion
Compatibil
Cost Reduction
Urgent
Change
Manufacturability
Other
Routine
Description of Problem
Initial menthol product delivery specifications were developed based on data from one
run in April 1988. Since that time several modifications to the product have been made.
SP-6M product.
Specifications need to be revised to reflect acceptable delivery ranges of the current
Proposed Solution
A sensory test was conducted on test products 8-033B and 8-035 to explore perceptual
effects related to delivery extremes. From that test specification limits were
recommended. Targets were verified on a recent menthol production run (8-049)
using final production materials. The recommended specifications are attached.
51482 5049
Signature
Title
Date
Accepted for
Investigation
Final Disposition
Approved
Rejected
SPEC HOLDER
Cailin . Carpenter
DATE
Action Taken or Reason for Rejection
8:24.8x
PRODUCT REVIEW COMMITTEE
RDP
KAN
PRODUCT CHANGE COMMITTEE
Proposed Effective Date:
PRODUCT ACCEPTANCE COMMITTEE
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
8-24-88
Models
4 of 4 columns · click a model to add or remove it

Ours

Exact match

8-24-88

7 characters8 tokens

Aux 2015

Partial · ANLS 0.57

8:24.8x

7 characters8 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

RJRT REQUEST FOR CHANGE F-1001 7/88

Question:

What is the date of preparation ?

R&D New Product Technology - BGTC Brand/Assembly/Material Affected Menthol Product Delivery Specification Type of

200 characters64 tokens