The Genealogy of Eliza

Automated excerpt

The current corpus contains 27 specimens spanning 1965–2025 in eight languages, described by 13 binary features. The current features therefore capture general pattern-matching architecture rather than anything specifically conversational, and better conversational features are needed. A language model proposed structural and behavioral features for a first program.

Selected automatically from source text; not independently written or fact-checked. Read the original for full context.