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.

翻译：语用学是自然语言的核心，使说话者能够利用省略和照应等结构高效沟通，在缩短话语的同时不损失意义。这些结构要求听者解读歧义形式——如代词——并推断说话者的意图——即该代词所指的对象。尽管可能引入歧义，照应在人类语言中普遍存在。为深入理解自然语言中照应结构的起源，我们探究在解决交际任务的人工神经网络之间是否能涌现类似结构。研究表明：首先，尽管歧义可能增加，含有照应结构的语言仍能被神经模型学习；其次，照应结构在模型之间"自然"涌现，无需额外约束；最后，对说话者施加明确的效率压力会提高这类结构的出现频率。我们得出结论：某些语用结构无需明确的效率压力即可在神经网络间直接涌现，但说话者与听者之间的竞争需求会制约其涌现程度与性质。