论文标题
角色级表示即使在伯特时代也可以改善基于DRS的语义解析
Character-level Representations Improve DRS-based Semantic Parsing Even in the Age of BERT
论文作者
论文摘要
我们结合了角色级别和上下文语言模型表示，以提高话语表示结构解析的性能。角色表示可以轻松地以一个编码器或完全独立的编码器中的顺序到序列模型添加，对不同语言模型，语言和数据集的改进都可以进行。对于英语，这些改进要比添加单个语言信息来源或添加非上下文嵌入更大。一种基于语义标签的新分析方法表明，字符级表示改善了选定语义现象的一部分的性能。
We combine character-level and contextual language model representations to improve performance on Discourse Representation Structure parsing. Character representations can easily be added in a sequence-to-sequence model in either one encoder or as a fully separate encoder, with improvements that are robust to different language models, languages and data sets. For English, these improvements are larger than adding individual sources of linguistic information or adding non-contextual embeddings. A new method of analysis based on semantic tags demonstrates that the character-level representations improve performance across a subset of selected semantic phenomena.
