Hallucination Types in Language Models: A Complete Guide › Visualizing the Taxonomy

Michael BrenndoerferMarch 18, 20262 min read
Language AI HandbookAll chapters
Part 56 of 67Hallucination and Factuality
Chapter 429Hallucination Types in Language Models: A Complete Guide › Visualizing the Taxonomy
Reading section 9 of 15 · Visualizing the Taxonomy · 2 minSection 9 of 15 · 2 min

Book contents

Language AI Handbook67 parts · 513 chaptersView the whole book →
Active chapter 429 of 513

Section 9 of 15 · 2 min read

Hallucination Types in Language Models: A Complete Guide › Visualizing the TaxonomyOne section · Sections in this chapter9 / 15 ↓
Sections in this chapter15
Reading tools Definitions & inline links

Choose your expertise level to adjust how many terms are explained. Beginners see more tooltips, experts see fewer to maintain reading flow. Hover over underlined terms for instant definitions.

Article links

Make inline references clickable

Let's visualize the full taxonomy of hallucination types to consolidate these concepts before diving into code.

Out[5]:
Visualization
Hierarchical taxonomy diagram of hallucination types with two primary branches and five leaf categories.
A taxonomy of language model hallucination types organized by their detection requirements. Source-grounded hallucinations (intrinsic and extrinsic) are evaluated by comparing output to a provided source document. Open-domain hallucinations (factual errors and fabrication) require external world knowledge for verification. Inconsistency spans both settings and is detectable from the output alone without any external reference.