TAAFT
Free mode
100% free
Freemium
Free Trial
Deals
Create tool
Glossary
Natural Language Processing
Perplexity

Perplexity

[pərˈplɛksɪti]
Natural Language Processing
Last updated: December 9, 2024

Definition

A measurement of how well a probability model predicts a sample, commonly used in language models.

Detailed Explanation

Perplexity is the exponential of the average negative log-likelihood per word. Lower perplexity indicates the model is better at predicting the sample. It can be interpreted as the weighted average branching factor of a language model.

Use Cases

Used in evaluating language models, speech recognition systems, and text generation models.

Related Terms

Named Entity Recognition

A technique that identifies and classifies named entities (people, organizations, locations, etc.) in text.

Natural Language Processing

Natural Language Processing

AI that enables computers to understand and interpret human language.

Natural Language Processing

Coreference Resolution

The task of finding all expressions that refer to the same entity in a text.

Natural Language Processing