Abstract screening is the first pass: you read each citation's title, abstract, and keywords and
decide whether it's potentially relevant. VADRR serves citations one at a time, ordered by the
machine-learning model so the most promising appear first.
Who can do this: any reviewer assigned to the screening.

The citation title sits in the top bar; the abstract and keywords fill the reading pane; the
decision buttons are at the bottom and the reasoning menu is on the right.
At the bottom of the screen are three buttons:
| Button | Meaning | Label value |
|---|---|---|
| Green ✓ | Accept (include) | 1 |
| Purple | Maybe | 0 |
| Red ✗ | Reject (exclude) | -1 |
Click one to record your decision. A short confirmation (Accepted / Maybe / Rejected) appears,
and VADRR automatically advances to the next citation.
If the screening requires a reason, tag, note, or form for that decision, VADRR won't save it until
you provide it — you'll see "This label requires additional information. Please see 'Required for
labels'." and the missing item is flagged in red.
On a wide screen the reasoning menu sits on the right; on a small screen tap the ≡ button to
open it. It has three sections:
Each section shows "Required for labels: Yes / Maybe / No" when that information is mandatory for a
given decision.
You can highlight important words to scan abstracts faster. Your project Leader may have already done so. These terms will help you identify easier whether the abstract fits your projects scope. To set ths up:
Matching terms are then highlighted in color wherever they appear in the title, abstract, and
keywords.
When the pool is exhausted you'll see "All citations screened" with a prompt to use Your Labels
to review or edit past decisions.