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 · pagenrcj0037_8.png

Who is the R&D customer for the 2nd project in the list?

OCR text of the page · 2,902 characters; the scanned image itself is not published
Appendix IIA
Methods Development Two-Year Project Plan
R&D
PROJECTS COMMITTED TO DO
DURATION PROJECT
CUSTOMER
PRIORITY
MONTHS
LEADER
1. CV HCN/Acetaldehyde in Smoke
SCM
2. MACON QA Flavor/Concentrate Methods
RHH
1.0
1.0
NMD2
KAF
1.0
DCH,TVB
3. Pack Seal vs MTVR
FY
4. RAA Lancaster Glycerin
DLG
7.0
5. RAA HCN/Acetaldehyde
SCM
1.0
6. Eval. ION TRACK Pack Seal Tester
KAF
1.0
4
/LG
4
VLG
1.0
4
FY
7. RAA Lancaster Sugars
OLG
B. Thumper Coal Retention
AM
1.0
VLG
9. RAA Lancaster Phosphate
DLG
1.0
FY
10. HAA Lancaster Ammonia
DLG
1.0
VLG
1 1. (C) HPLC For Licorice On Wet Snuff
KMW
12. (C) Nicotine for STP
RRB
1.0
VLG
FY
13. (V) Oven Moisture For RECON/Oven Cal Procedure
KMW
1.0
ive
14. Automate Sample Prep for GC Smoke Machine
KMW
FY
15. Replace Shazbot
NMD2
KMW
16. (C,VI Oven Moisture for STP
KMW
NMD2
17. Revise Hot Water Extractables
KMW
TVB
KMW
IVE
78. RAA Ammonia on Tobacco Blends, Grades, Recon
RHHIDWCI
VLG
19. Quantitate HFP for Tobacco
NMD1
20. Optimize/Standardize In-Cig. Obj. Stem Method
KRK
21. Quantitate HFP for tobacco
SCM
2.0
FY
NMD 1
22. Finish Quantitativef DS Scan for Tobacco
DLG
2.0
2.0
DCH
23. Automate HFP Sample Prep
RHHIDWC)
2.0
24. RAA Nitrite on Tobacco/Snuff
RRB
2.0
NMD2
25. (C) RAA Urea, Glycerin (RECON)
VLG
KMW
26. Automated pH for Tobacco/RECON
2. 1
VLG
XMW
27. Automated RAA Sample Prep
2.2
NMD2
KMW
28. HF Automated Sample Prep Feasibility for Megabore
2.3
KMW
RHH(DWC)
2.4
DCH
29. Automate DS Scan Sample Prep
DCH
30. Standardize WTS OBJ. Stem Method
31. (C) Methods Physical Properties Tobacco (TA, Instron)
KRK
3.0
DCH
3.0
NMD2
32. Re-evaluate CTS-400 Performance
DLC
AM
FY
33. Robotic Recon Base Wt.
KMW
3.0
3.0
NMD 1
34. DAP in MGC-817
DCH
AM
3.1
35. Develop Better Cut Width Measurement Method
KRK
TVE
4.0
36. (V)Oven Moisture For RECON/Over Cal Procedure
DLG
NMD2
37. RAA Maltose on Tobacco
4.0
AM
4.0
FY
38. (V) Triglycer/Butterfat on Tobacco
4.0
VLG
39. RAA Ammonia on Tobacco Blends, Grades, RECON
SCM
TVB
RRB
4.0
40. RAA PG on Tobacco
AM
41. RAA Citrate on Paper
AM
4.1
VLG
NMD3
42. GC Method for Snuff Flavors
ARB
5.0
NMD3
5.0
TVB
43. Automate/Quantitate Teabag
RHHIDWC)
44, RAA Phosphate on Paper
KMW
5.0
NAADANNAANNNNA DAD NANODANTAWA DATA WON WNA W - A ON - NPACT
NMD1
45. (Vi Megabore Men/Tip/Tob Sec
5.3
AM
NMD3
6.0
46. (C) HPLC For Licorice On Wet Snuff
RAB
TVB
47. Smoke Ammonia
6.0
FY
AM
48. (V) Puff by Puff
6.1
AM
NMD 1
49. RAA Sugars in Casings [F, G,S,M)
6.3
NMD 1
50. (C V] Oven Moisture for STP
AM
7.0
NMD3
RRE
51. Finish HPLC Lic/Coc for Cigarette Tobacco
TVB
52. (C) RAA Urea, Glycerin(RECON)
AM
7.0
8.0
TVB
153. (V)Smoke pH
RR8
54. (V) Integrate Argon Blend Into Blend Sep
RRB
1.0
VLG
14.0
RRB
15.0
NMD
55. Sand/inorganics in Tobacco
DCH
RRB
17.0
NMD2
* V = Validation
NMD1 = STH Replacement
C = Current
NMD2 = Robotics/Analytical Chemist
NMD3 = VLG Replacement
570363436
Source: https://www.industrydocuments.ucsf.edu/docs/nrcj0037
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
RHH
Models
4 of 4 columns · click a model to add or remove it

Ours

Wrong

MACON

5 characters3 tokens

Aux 2015

Wrong

The question asks: "Who is the R&D customer for the 2nd project in the list?"

Looking at the provided text, the list of projects is:

  1. CV HCN/Acetaldehyde in Smoke
  2. MACON QA Flavor/Concentrate Methods
  3. Pack Seal vs
220 characters64 tokens

PiT-FT 2015

Wrong

Empty response.

0 characters

ChronoGPT 2015

Wrong

Input:

Question:

Who is the R&D customer for the 2nd project in the list?

Question:

Who is the R&D customer for the 2nd project in the list?

Question:

Who is

179 characters64 tokens