Syntax Aware LSTM Model for Chinese Semantic Role Labeling

TL;DR

This paper introduces a syntax-aware LSTM model for Chinese semantic role labeling that integrates dependency parsing information directly into the model architecture, leading to significant performance improvements.

Contribution

The paper presents a novel SA-LSTM model that incorporates dependency parsing information directly into the architecture, avoiding feature engineering.

Findings

SA-LSTM outperforms previous models on CPB 1.0 dataset

Model architecture benefits more from parsing information than feature engineering

Significant statistical improvement over state-of-the-art methods

Abstract

As for semantic role labeling (SRL) task, when it comes to utilizing parsing information, both traditional methods and recent recurrent neural network (RNN) based methods use the feature engineering way. In this paper, we propose Syntax Aware Long Short Time Memory(SA-LSTM). The structure of SA-LSTM modifies according to dependency parsing information in order to model parsing information directly in an architecture engineering way instead of feature engineering way. We experimentally demonstrate that SA-LSTM gains more improvement from the model architecture. Furthermore, SA-LSTM outperforms the state-of-the-art on CPB 1.0 significantly according to Student t-test ($p<0.05$).