Establishing Strong Baselines for the New Decade: Sequence Tagging, Syntactic and Semantic Parsing with BERT

Han He, Jinho D. Choi

Best Student Paper Candidate


Abstract

This paper presents new state-of-the-art models for three tasks, part-of-speech tagging, syntactic parsing, and semantic parsing, using the cutting-edge contextualized embedding framework known as BERT. For each task, we first replicate and simplify the current state-of-theart approach to enhance its model efficiency. We then evaluate our simplified approaches on those three tasks using token embeddings generated by BERT. 12 datasets in both English and Chinese are used for our experiments. The BERT models outperform the previously bestperforming models by 2.5% on average (7.5% for the most significant case). Moreover, an indepth analysis on the impact of BERT embeddings is provided using self-attention, which helps understanding in this rich yet representation. All models and source codes are available in public so that researchers can improve upon and utilize them to establish strong baselines for the next decade.

Venue / Year

Proceedings of the International Florida Artificial Intelligence Research Society Conference (FLAIRS) / 2020

Links

Anthology | Paper | BibTeX