Dialogue Games for Benchmarking Language Understanding: Motivation, Taxonomy, Strategy
This addresses the challenge of evaluating language understanding in AI systems, offering a foundational approach that could impact benchmarking in natural language processing.
The paper tackles the problem of measuring artificial language understanding by proposing Dialogue Games as a systematic test embedded in practice, arguing that formal tests alone are insufficient and that these games provide construct validity and strategic guidance for development.
How does one measure "ability to understand language"? If it is a person's ability that is being measured, this is a question that almost never poses itself in an unqualified manner: Whatever formal test is applied, it takes place on the background of the person's language use in daily social practice, and what is measured is a specialised variety of language understanding (e.g., of a second language; or of written, technical language). Computer programs do not have this background. What does that mean for the applicability of formal tests of language understanding? I argue that such tests need to be complemented with tests of language use embedded in a practice, to arrive at a more comprehensive evaluation of "artificial language understanding". To do such tests systematically, I propose to use "Dialogue Games" -- constructed activities that provide a situational embedding for language use. I describe a taxonomy of Dialogue Game types, linked to a model of underlying capabilites that are tested, and thereby giving an argument for the \emph{construct validity} of the test. I close with showing how the internal structure of the taxonomy suggests an ordering from more specialised to more general situational language understanding, which potentially can provide some strategic guidance for development in this field.