Holistic Exploration on Universal Decompositional Semantic Parsing: Architecture, Data Augmentation, and LLM Paradigm

TL;DR

This paper presents a comprehensive study on Universal Decompositional Semantic Parsing, introducing a cascade model, data augmentation techniques, and evaluating the use of ChatGPT, achieving improved performance and efficiency.

Contribution

The paper proposes a novel cascade architecture for UDS parsing, explores data augmentation methods, and assesses ChatGPT's effectiveness, advancing the state-of-the-art in semantic parsing.

Findings

The cascade model outperforms previous models in accuracy and speed.

Data augmentation techniques further improve parsing performance.

ChatGPT excels at attribute parsing but has limitations in relation parsing.

Abstract

In this paper, we conduct a holistic exploration of the Universal Decompositional Semantic (UDS) Parsing. We first introduce a cascade model for UDS parsing that decomposes the complex parsing task into semantically appropriate subtasks. Our approach outperforms the prior models, while significantly reducing inference time. We also incorporate syntactic information and further optimized the architecture. Besides, different ways for data augmentation are explored, which further improve the UDS Parsing. Lastly, we conduct experiments to investigate the efficacy of ChatGPT in handling the UDS task, revealing that it excels in attribute parsing but struggles in relation parsing, and using ChatGPT for data augmentation yields suboptimal results. Our code is available at https://github.com/hexuandeng/HExp4UDS.