"Not not bad" is not "bad": A distributional account of negation

TL;DR

This paper proposes a tripartite continuous vector space model to better capture negation in distributional semantics, addressing limitations of current models in representing logical operations.

Contribution

It introduces a novel tripartite vector space formulation for semantics, enabling formal compositional negation within distributional models.

Findings

The tripartite model improves negation representation.

It provides a formal framework for compositional semantics.

Enhances the logical capabilities of distributional models.

Abstract

With the increasing empirical success of distributional models of compositional semantics, it is timely to consider the types of textual logic that such models are capable of capturing. In this paper, we address shortcomings in the ability of current models to capture logical operations such as negation. As a solution we propose a tripartite formulation for a continuous vector space representation of semantics and subsequently use this representation to develop a formal compositional notion of negation within such models.