Anaphoric Structure Emerges Between Neural Networks
This addresses the problem of understanding the origins of pragmatic language structures for researchers in computational linguistics and AI, though it is incremental as it builds on prior work on emergent communication.
The study investigated whether anaphoric structures, which shorten utterances without losing meaning, can emerge between neural networks trained on a communicative task, finding that such structures are learnable, emerge naturally without constraints, and increase with explicit efficiency pressure.
Pragmatics is core to natural language, enabling speakers to communicate efficiently with structures like ellipsis and anaphora that can shorten utterances without loss of meaning. These structures require a listener to interpret an ambiguous form - like a pronoun - and infer the speaker's intended meaning - who that pronoun refers to. Despite potential to introduce ambiguity, anaphora is ubiquitous across human language. In an effort to better understand the origins of anaphoric structure in natural language, we look to see if analogous structures can emerge between artificial neural networks trained to solve a communicative task. We show that: first, despite the potential for increased ambiguity, languages with anaphoric structures are learnable by neural models. Second, anaphoric structures emerge between models 'naturally' without need for additional constraints. Finally, introducing an explicit efficiency pressure on the speaker increases the prevalence of these structures. We conclude that certain pragmatic structures straightforwardly emerge between neural networks, without explicit efficiency pressures, but that the competing needs of speakers and listeners conditions the degree and nature of their emergence.