CSAGN: Conversational Structure Aware Graph Network for Conversational Semantic Role Labeling

TL;DR

This paper introduces CSAGN, a graph network model that explicitly encodes conversational structure and speaker information to improve semantic role labeling in dialogues, achieving significant performance gains.

Contribution

The paper proposes a novel conversational structure-aware graph network with multi-task learning for CSRL, addressing structural information handling and outperforming previous methods.

Findings

Significant improvement over previous baselines on benchmark datasets

Effective encoding of speaker-dependent information

Enhanced model performance with multi-task training

Abstract

Conversational semantic role labeling (CSRL) is believed to be a crucial step towards dialogue understanding. However, it remains a major challenge for existing CSRL parser to handle conversational structural information. In this paper, we present a simple and effective architecture for CSRL which aims to address this problem. Our model is based on a conversational structure-aware graph network which explicitly encodes the speaker dependent information. We also propose a multi-task learning method to further improve the model. Experimental results on benchmark datasets show that our model with our proposed training objectives significantly outperforms previous baselines.