论文标题
盖佩托（Geppetto）将意大利语变成语言模型
GePpeTto Carves Italian into a Language Model
论文作者
论文摘要
在过去的几年中，预训练的神经体系结构为几个NLP任务提供了令人印象深刻的改进。尽管如此，生成语言模型仍主要用于英语。我们开发了使用GPT-2体系结构构建的意大利语的第一个生成语言模型Geppetto。我们通过自动评估和基于人类的评估对Geppetto的质量进行了彻底的分析。自动评估包括（i）计算不同类型的困惑，以及（ii）对Geppetto的写作特征进行分析分析。我们发现Geppetto的生产是人类生产的一种盆景版本，其较短但又复杂的句子。人类评估是通过句子完成任务执行的，在该任务中，Geppetto的输出经常被判断为自然，并且比我们将其视为基线更简单的语言模型更接近原始人类文本。
In the last few years, pre-trained neural architectures have provided impressive improvements across several NLP tasks. Still, generative language models are available mainly for English. We develop GePpeTto, the first generative language model for Italian, built using the GPT-2 architecture. We provide a thorough analysis of GePpeTto's quality by means of both an automatic and a human-based evaluation. The automatic assessment consists in (i) calculating perplexity across different genres and (ii) a profiling analysis over GePpeTto's writing characteristics. We find that GePpeTto's production is a sort of bonsai version of human production, with shorter but yet complex sentences. Human evaluation is performed over a sentence completion task, where GePpeTto's output is judged as natural more often than not, and much closer to the original human texts than to a simpler language model which we take as baseline.
