(Re)construing Meaning in NLP
It introduces a theoretical concept from cognitive semantics to potentially enhance NLP models, but is incremental as it builds on prior work without new data or methods.
The paper addresses the absence of construal—how expression reflects conceptualization—in NLP meaning discussions, surveying its dimensions and showing how it could inform NLP work.
Human speakers have an extensive toolkit of ways to express themselves. In this paper, we engage with an idea largely absent from discussions of meaning in natural language understanding--namely, that the way something is expressed reflects different ways of conceptualizing or construing the information being conveyed. We first define this phenomenon more precisely, drawing on considerable prior work in theoretical cognitive semantics and psycholinguistics. We then survey some dimensions of construed meaning and show how insights from construal could inform theoretical and practical work in NLP.