Skip to content

Evaluation & Quality

Perplexity

A score describing how surprising a piece of text is to a language AI.

Example

AI gets a lower score for writing it finds easier to predict.

Why people use it

It measures how expected a passage is to a language AI system.

What you'll hear

“How surprising is this text to the system?”

What this means for you

Compare scores only when the text and scoring setup are compatible.

Can you control it?

No

No direct control. This describes a wider issue, concept or result rather than something you can simply switch on or off in a tool.

Common questions

Is perplexity here the name of the search product?
No. This entry concerns the checking score, which is a separate meaning.
Does low perplexity prove a statement is true?
No. Familiar or predictable wording can still contain false information.
Can scores from different word-splitting methods be compared directly?
Not reliably. Different ways of dividing text can change the scale of the score.

Related terms

Still have questions?

Up to 500 characters.

Ask LATHIC about AI. Relevant glossary entries may be included.

Your question, the glossary entries it matches, and a rotating pseudonymous identifier go to Microsoft Azure’s OpenAI service through Vercel AI Gateway to generate an answer. Zero retention and no training are required of the provider, and LATHIC does not save your question or answer. Privacy Notice