The cognitive mechanisms needed to account for the English past tense have long been a subject of debate in linguistics and cognitive science. Neural network models were proposed early on, but were shown to have clear flaws. Recently, however, Kirov and Cotterell (2018) showed that modern encoder-decoder (ED) models overcome many of these flaws. They also presented evidence that ED models demonstrate humanlike performance in a nonce-word task. Here, we look more closely at the behaviour of their model in this task. We find that (1) the model exhibits instability across multiple simulations in terms of its correlation with human data, and (2) even when results are aggregated across simulations (treating each simulation as an individual human participant), the fit to the human data is not strong---worse than an older rule-based model. These findings hold up through several alternative training regimes and evaluation measures. Although other neural architectures might do better, we conclude that there is still insufficient evidence to claim that neural nets are a good cognitive model for this task.
Mind Your Inflections! Improving NLP for Non-Standard Englishes with Base-Inflection Encoding
Salesforce AI Research（赛富时人工智能研究院）；National University of Singapore（新加坡国立大学）；Nanyang Technological University（南洋理工大学）；University of Illinois at Urbana-Champaign（伊利诺伊大学厄巴纳-香槟分校）
Unsupervised Inflection Generation Using Neural Language Modeling
Human Language Technologies Research Center, University of Bucharest（布加勒斯特大学人类语言技术研究中心）；GumGum（GumGum公司）；University of Bucharest, Faculty of Mathematics and Computer Science（布加勒斯特大学数学与计算机科学学院）
Inflecting when there's no majority: Limitations of encoder-decoder neural networks as cognitive models for German plurals
Institute for Language, Cognition and Computation（语言、认知与计算研究所）；School of Informatics（信息学院）；University of Edinburgh（爱丁堡大学）
人工神经网络能像人类说话者那样学会表示屈折形态并概括为新词吗？Kirov和Cotterell（2018）认为答案是肯定的：现代编码器-解码器（ED）架构在改变英语动词时学习类似人类的行为，例如将常规的过去时态形式-（e）d扩展到新单词。然而，他们的工作并没有解决Marcus et al.（1995）提出的批评：神经模型可能会学习扩展不是常规的，而是最常见的类-因此在德国数字变形等任务上失败，其中不常见的后缀如-s仍然可以有效地推广。为了研究这个问题，我们首先收集了一个新的数据集从德语（生产和评级的复数形式的新名词），旨在避免信息来源不可用的ED模型。说话人数据显示出高变异性，两个后缀表明“规则”的行为，出现更频繁的语音非典型输入。编码器-解码器模型确实概括了最频繁产生的复数类，但没有显示出这些其他复数标记的类似人类的可变性或“规则”扩展。我们的结论是，现代神经模型可能仍在努力解决少数群体的泛化问题。
SIGMORPHON 2020 Shared Task 0: Typologically Diverse Morphological Inflection
University of Melbourne（墨尔本大学）；University of Cambridge（剑桥大学）；Johns Hopkins University（约翰斯·霍普金斯大学）；Higher School of Economics（高等经济学院）；Moscow State University（莫斯科国立大学）；Karelian Research Centre（卡累利阿研究中心）；Indiana University（印第安纳大学）；Google AI（谷歌人工智能）；University of British Columbia（不列颠哥伦比亚大学）；Facebook AI Research（Facebook人工智能研究院）；Carnegie Mellon University（卡内基梅隆大学）；University of Louisville（路易斯维尔大学）；University of York（约克大学）；ETH Zürich（苏黎世联邦理工学院）；University of Colorado Boulder（科罗拉多大学博尔德分校）
