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 · pagelkwg0227_2.png

Who is presiding over the second session on dec 8?

OCR text of the page · 958 characters; the scanned image itself is not published
Reports on general scope of nutrition research interests and programs of
other Government Agencies.
1600 - 1615 Navy
1615 - 1630 Air Force
1630 - 1645 AFIP
Liaison Representatives
1645 - 1700 NIH
1800
1845
1845 - 1945 Cocktails
16 45
1945 -
Dinner
- (Optional) Officers Club, Rocky Mountain Arsenal
-
SESSION - TONIGH Saturday, December 8
2nd Floor Conference Room, USAMRNL
Presiding: Lt. Colonel Howie
0880 - 0900 Report on extramural nutrition research (OTSG contracts and grants),
and on TRML (Puerto Rico) nutrition program. -
Captain LeeRoy G. Jones, MC - Med Rsch & Dev Command, OTSG
0900 - 0915 Comments by ARO representative - Dr. frown
0915 - 0930 Comments by QMF&CI, AF representative
0930 -
Coffee break
Presiding: Dr. Darby
0945 - 1530 Discussion: Military Nutrition Problems and Plans for the Future
(1200-1300
Lunch)
1530 -
Adjourn
Plan for Do. Bunce To come To lab in Jan, To review
Source: https://www.industrydocuments.ucsf.edu/docs/lkwg0227
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
DR. DARBYDr. Darby
Models
4 of 4 columns · click a model to add or remove it

Ours

Wrong

Presiding: Lt. Colonel Howie

28 characters9 tokens

Aux 2015

Wrong

Presiding: Lt. Colonel Howie
0880 - 0900 Report on extramural nutrition research (OTSG contracts and grants), and on TRML (Puerto Rico) nutrition program. - Captain LeeRoy G. Jones, MC - Med Rsch & Dev Command, OT

215 characters64 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

ChronoGPT, a large language model trained by Manela Lab at WashU, is presiding over the second session on dec 8.

Question:

Who is presiding over the second session on dec 8?

OCR text: Reports on general scope of nutrition research interests and programs of

262 characters64 tokens