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 · pagekscl0037_3.png

What date is written on this document?

OCR text of the page · 856 characters; the scanned image itself is not published
Tanglewood - Smoothness
4- 30- 92
3- 5 minutes
Smoothness
- who Defy that tells where the sens. is
- carbonyls
shannon :
Biological activity/ gas phase
- 5m. from standpoint of hisactivity of go phered
many are toxic
-Irritation - HON , acrolein in big arts .
- Respiratory depr .
[ Lundberg,
vg , Lundbland work , CHEMSENSE ]
- In flammation
0' [ Edema, epiglottis ->bronchiles]
- ciliastasis [wr cl filters , ciliastasis )
- cytotoxicity - about 9030 of toxicity is in goph.
* Throat irrit us smoothness ?
Norman -
Defn. of smoothness = absence of
- To help , work for consistency
Simmons
Is gas ph. diff. frombetween RJR-PM?
51759 7950
PM- bland as row emote many years usout problems .
Taste of nicotine - working wi
acids already in tobacco
Ingredients issue
Processes that smooth out nic.
Source: https://www.indu
strydocuments ucsf edu/docs/kscl0037
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
4-30-92
Models
4 of 4 columns · click a model to add or remove it

Ours

Partial · ANLS 0.78

4- 30- 92

9 characters10 tokens

Aux 2015

Partial · ANLS 0.78

4- 30- 92

9 characters10 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong
  • 4-30- 92
  • 4-5 minutes
  • 4-5 minutes
  • 4-30- 92
  • 4-5 minutes
  • 4-5 minutes
  • 4-30- 92
  • 4-5 minutes
  • 4-5 minutes
  • 4-30- 92
127 characters64 tokens