How verified citations work: the four verdicts

CiteDash AI's core promise is that no citation is ever invented and no claim reaches you unchecked. This guide explains the machinery behind that promise so you know exactly what a green checkmark does (and does not) mean.

Citations are database objects, not text

In most AI tools, a citation is a string the model wrote, which is why invented references happen. In CiteDash, a citation is a row that links your sentence to a real paper record. A generated citation that doesn't resolve to a real paper is rejected before display. There is no path around this.

What verification checks

For every cited sentence, the Fact Checker retrieves the relevant passages from the cited paper's stored full text and judges whether they support the claim. It never uses the paper's abstract as evidence and never falls back on the model's memory of the paper. If the full text isn't held, the honest answer is 'unverified', and that's what you get.

The four verdicts

• Supported: the source's full text backs the claim as written.
• Partial: the source supports part of the claim; check the qualifier you added.
• Unsupported: the source does not back this claim. Rewrite it, cite a different source, or remove it.
• Unverified: the claim hasn't been checked yet, or the cited paper has no stored full text. Add the PDF and re-verify.

Fixing verdicts in practice

In Thesis Editor, per-paragraph indicators show grounding at a glance and one click re-verifies the whole document. The most common fix is narrowing an overclaim: "X causes Y" often becomes supported as "X was associated with Y in two RCTs". The second most common is promoting a held reference to full text so verification has something to read.

Retraction is checked continuously as well. A paper retracted after you cited it is flagged in the Reference Manager and again at compile time, with the offending citation named.