EDR Review Template

Dataset Integrity

• Expected responses:
• Present responses:
• Missing responses:
• Malformed responses:
• Comparator version:
• Run ID:

Direct Observations

List only observations directly supported by comparator outputs.

Instrument Limitations

• parsing issues:
• comparator limitations:
• dataset weaknesses:
• provider-specific concerns:

Explainability Questions

• Is the disagreement lexical or structural?
• Does the difference alter the backbone?
• Is one response compressed or expanded?
• Did a model add unsupported structure?
• Did recognition affect extraction?
• Did domain vocabulary affect the result?
• Is the difference scientifically relevant?
• Does it expose a comparator limitation?
• Should it affect hypothesis review?
• Is additional evidence required?

Interpretation

Supported by evidence:

Reviewer inference:

Hypothesis Impact

For each referenced hypothesis include:

• Supporting observations:
• Challenging observations:
• Unknowns:

Do not assign confidence automatically.

Recommended Follow-up

• additional experiment needed
• replication recommended
• no further work
• insufficient evidence

Review Outcome

• Accepted
• Accepted with Notes
• Needs Revision
• Deferred