prepare Trott et al. (2020)

[BarbaraMcG]

(Re)construing Meaning in NLP Sean Trott | Tiago Timponi Torrent | Nancy Chang | Nathan Schneider https://www.aclweb.org/anthology/2020.acl-main.462.pdf

1. What is this paper about?
2. Is it relevant to our project? If so, why and how?
3. What could we use from this work in our project?
4. Add some text about it to Overleaf
5. Plan experiments (if appropriate)

[GiorgiatolfoBL]

The paper addresses the different ways that information is conceptualised and constructed by language. It claims that this is a problem largely absent from discussions of meaning in NLP and offers in this respect a threefold approach to the theme: definition of linguistic elements, psycholinguistic evidence and relative NLP research. Meaning and construal (= the dynamic process of meaning creation) are closely linked. How much and what kind of construal is embedded in the meaning creation process can vary a lot. Among the various dimensions of construed meaning we can note: perspective (vantage point), prominence (focus), resolution (detailed>abstract), configuration (internal-structural properties of entities), metaphor. From an NLP perspective these can be approached studying active/passive constructions, aspectual composition, count/mass etc. Construal poses a challenge to NLP and it should not be overseen.

The paper raises an interest point and it is important as far as it regards the need to acknowledge the limits of close reading. In acknowledging the difficulty that studying a slippery thing like the construal it advocated for a a negotiation between close and distant reading, one that tries to take into consideration as much as possible the context in which an information has been created and conveyed.

It is important for the project in terms of awareness but I don't think it provides examples or ideas to implement.