Structural Information Preserving for Graph-to-Text Generation
This work addresses a crucial defect in graph-to-text generation for NLP applications, though it is incremental as it builds on existing methods with enhanced training signals.
The paper tackles the problem of graph-to-text generation models dropping core structural information by introducing autoencoding losses that preserve different aspects of input graphs, resulting in improved performance over a state-of-the-art baseline on two benchmarks.
The task of graph-to-text generation aims at producing sentences that preserve the meaning of input graphs. As a crucial defect, the current state-of-the-art models may mess up or even drop the core structural information of input graphs when generating outputs. We propose to tackle this problem by leveraging richer training signals that can guide our model for preserving input information. In particular, we introduce two types of autoencoding losses, each individually focusing on different aspects (a.k.a. views) of input graphs. The losses are then back-propagated to better calibrate our model via multi-task training. Experiments on two benchmarks for graph-to-text generation show the effectiveness of our approach over a state-of-the-art baseline. Our code is available at \url{http://github.com/Soistesimmer/AMR-multiview}.