Sequence Labeling Parsing by Learning Across Representations
This work addresses parsing efficiency and accuracy for NLP researchers, but it is incremental as it builds on existing multitask learning and sequence labeling methods.
The paper tackles the problem of improving syntactic parsing by learning across constituency and dependency representations using multitask learning, resulting in average performance gains of 1.14 F1 points for constituency parsing and 0.62 UAS points for dependency parsing.
We use parsing as sequence labeling as a common framework to learn across constituency and dependency syntactic abstractions. To do so, we cast the problem as multitask learning (MTL). First, we show that adding a parsing paradigm as an auxiliary loss consistently improves the performance on the other paradigm. Secondly, we explore an MTL sequence labeling model that parses both representations, at almost no cost in terms of performance and speed. The results across the board show that on average MTL models with auxiliary losses for constituency parsing outperform single-task ones by 1.14 F1 points, and for dependency parsing by 0.62 UAS points.