Aiplorer Summary Reference Pack | Version 1 | Comparison worksheet
Blank local worksheet. No Aiplorer shortlist data or private notes are included.
Keep completed files in an approved location; do not enter sensitive source data.

QUESTION FOR THIS COMPARISON
[What specific change or unresolved question makes another run useful?]

RUN A
Date and timezone:
Tool / account type / visible model or settings (use "not shown" when unknown):
Input pack version and any deviations:
Exact instruction and input file location:
Original output file location:

S-01 outcome (meets these checks / needs correction / could not test):
Source IDs and observed issue or evidence:
S-02 outcome:
Source IDs and observed issue or evidence:
S-03 outcome:
Source IDs and observed issue or evidence:
Repair instructions and separate repaired output location, if any:

RUN B
Date and timezone:
Tool / account type / visible model or settings (use "not shown" when unknown):
What changed since A? What could not be held constant?
Input pack version and any deviations:
Exact instruction and input file location:
Original output file location:

S-01 outcome (meets these checks / needs correction / could not test):
Source IDs and observed issue or evidence:
S-02 outcome:
Source IDs and observed issue or evidence:
S-03 outcome:
Source IDs and observed issue or evidence:
Repair instructions and separate repaired output location, if any:

DECISION AND NEXT VISIT
Which specific claim handling improved, regressed or stayed unresolved?
What can this small comparison NOT establish?
Decision for this workflow (not a general recommendation):
Next question or relevant change to watch for:
Non-sensitive one-line reminder for My Shortlist:

Do not infer causation if multiple conditions changed. Do not total these
outcomes into a ranking. A completed check is not necessarily a passing result.