Evaluation UI: Update Results screen #6
Description
Is your feature request related to a problem?
The evaluation screen shows the question, ground truth, and answer as separate columns. But the question and ground truth stay the same, while the answer changes with each run. This creates visual clutter and makes it harder to quickly see what’s fixed context versus what’s actually changing.
Describe the solution you’d like
Group the question and ground truth together at the start of each row, with a subtle grey background to show they’re fixed reference data. Show the generated answer separately after that, so it’s clearly visible and easy to compare across runs.
Screenshot
