Dr Simon Wells

sesquipedalia verba

New Paper at CMNA'22

My latest paper, titled “Reconsidering RepStat Rules in Dialectical Games”, will be presented at the upcoming 22nd Annual International Workshop on Computational Models of Natural Argument (CMNA'22).

The paper is joint work with Mark Snaith from RGU in which we investigate rules governing repetition of statements within dialectical games in the context of conversational AI. We propose that one solution is to enable the participants to engage according to different sets of rules. This way the rules for agents generally need to be more restrictive and “formal dialectics” in order to maintain requirements for computational tractability, and the rules for people need to be more flexible and coincide with “descriptive dialectics”, to enable people to engage in ways that match natural interaction.

Prohibition of repeated statements has benefits for the tractability and predictability of dialogues carried out by machines, but doesn’t match the real world behaviour of people. This gap between human and machine behaviour leads to problems when formal dialectical systems are applied in conversational AI contexts. However, the problem of handling statement repetition gives insight into wider issues that stem partly from the historical focus on formal dialectics to the near exclusion of descriptive dialectics. In this paper we consider the problem of balancing the needs of machines versus those of human participants through the consideration of both descriptive and formal dialectics integrated within a single overarching dialectical system. We describe how this approach can be supported through minimal extension of the Dialogue Game Description Language.