arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 2103.02190cs.CLcs.LG
收藏

一种具有二阶注意力的迭代上下文化算法

An Iterative Contextualization Algorithm with Second-Order Attention

Université du Québec à Montréal(魁北克大学蒙特利尔分校)

Diego Maupomé, Marie-Jean Meurs

摘要:

将组成句子的单词的表示组合成一个有凝聚力的整体是困难的,因为它需要考虑单词的顺序,并建立单词之间的相互关系。我们提出的解决方案包括迭代地调整上下文。我们的算法从上下文的一个可能错误的值开始,并根据手头的令牌调整这个值。为了实现这一点,单词的表示被构建为将它们的符号嵌入与位置编码组合成单个向量。然后,该算法迭代加权和聚合这些向量使用我们的新的二阶注意力机制。我们的模型在几个著名的文本分类任务中报告了强大的结果。

英文摘要:

Combining the representations of the words that make up a sentence into a cohesive whole is difficult, since it needs to account for the order of words, and to establish how the words present relate to each other. The solution we propose consists in iteratively adjusting the context. Our algorithm starts with a presumably erroneous value of the context, and adjusts this value with respect to the tokens at hand. In order to achieve this, representations of words are built combining their symbolic embedding with a positional encoding into single vectors. The algorithm then iteratively weighs and aggregates these vectors using our novel second-order attention mechanism. Our models report strong results in several well-known text classification tasks.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑