Abstract
Temporal expressions in text play a significant role in language understanding and correctly identifying them is fundamental to various retrieval and natural language processing systems. Previous works have slowly shifted from rule-based to neural architectures, capable of tagging expressions with higher accuracy. However, neural models can not yet distinguish between different expression types at the same level as their rule-based counterparts. n this work, we aim to identify the most suitable transformer architecture for joint temporal tagging and type classification, as well as, investigating the effect of semi-supervised training on the performance of these systems. After studying variants of token classification and encoder-decoder architectures, we ultimately present a transformer encoder-decoder model using RoBERTa language model as our best performing system. By supplementing training resources with weakly labeled data from rule-based systems, our model surpasses previous works in temporal tagging and type classification, especially on rare classes. Additionally, we make the code and pre-trained experiment publicly available
Abstract (translated)
URL
https://arxiv.org/abs/2109.14927
