Related Experiment Video
Updated: Apr 20, 2026

Augmenting Large Language Models via Vector Embeddings to Improve Domain-Specific Responsiveness
Published on: December 6, 2024
Two interpretive systems for natural language?
1University of Massachusetts, Amherst, MA, USA, lyn@linguist.umass.edu.
Abstract:
It is proposed that humans have available to them two systems for interpreting natural language. One system is familiar from formal semantics. It is a type based system that pairs a syntactic form with its interpretation using grammatical rules of composition. This system delivers both plausible and implausible meanings. The other proposed system is one that uses the grammar together with knowledge of how the human production system works. It is token based and only delivers plausible meanings, including meanings based on a repaired input when the input might have been produced as a speech error.
Related Concept Videos
Language and Cognition
Automatic Processing and Automatic Social Behavior
Components of Language
Information Processing Approach
Language
Corballis and Suddendorf (2007) and Tomasello and Rakoczy (2003) highlight the role of language in...
Natural and Artificial Concepts