BERT got a Date: Introducing Transformers to Temporal Tagging
This work addresses the challenge of accurately identifying temporal expressions for natural language processing systems, representing an incremental improvement over existing neural models.
The paper tackled the problem of joint temporal tagging and type classification in text by identifying the most suitable transformer architecture and using semi-supervised training with weakly labeled data. The result was a RoBERTa-based encoder-decoder model that surpassed previous works, especially on rare classes.
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. In 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. Based on our study of token classification variants and encoder-decoder architectures, we present a transformer encoder-decoder model using the 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. Our code and pre-trained experiments are available at: https://github.com/satya77/Transformer_Temporal_Tagger