Composing Conversational Negation

TL;DR

This paper extends a framework for modeling negation in natural language by incorporating compositional structure and context, enabling more accurate representation of negated sentences and evolving word meanings.

Contribution

It introduces a method to compose negations of individual words within the DisCoCirc framework, capturing sentence-level negation and dynamic word meanings.

Findings

Successfully models negation at the sentence level

Handles evolving word meanings in context

Enhances previous word-level negation frameworks

Abstract

Negation in natural language does not follow Boolean logic and is therefore inherently difficult to model. In particular, it takes into account the broader understanding of what is being negated. In previous work, we proposed a framework for the negation of words that accounts for 'worldly context'. This paper extends that proposal now accounting for the compositional structure inherent in language within the DisCoCirc framework. We compose the negations of single words to capture the negation of sentences. We also describe how to model the negation of words whose meanings evolve in the text.