Stage-wise Fine-tuning for Graph-to-Text Generation
TL
TL;DR
This paper proposes a structured graph-to-text model with a two-step fine-tuning mechanism which first fine-tunes model on Wikipedia before adapting to the graph- to-text generation, and proposes a novel tree-level embedding method to capture the inter-dependency structures of the input graph.
Abstract
Graph-to-text generation has benefited from pre-trained language models (PLMs) in achieving better performance than structured graph encoders. However, they fail to fully utilize the structure information of the input graph. In this paper, we aim to further improve the performance of the pre-trained language model by proposing a structured graph-to-text model with a two-step fine-tuning mechanism which first fine-tunes model on Wikipedia before adapting to the graph-to-text generation. In addition to using the traditional token and position embeddings to encode the knowledge graph (KG), we propose a novel tree-level embedding method to capture the inter-dependency structures of the input graph. This new approach has significantly improved the performance of all text generation metrics for the English WebNLG 2017 dataset.
References41 items
1
Transformers: State-of-the-Art Natural Language Processing
2
Investigating Pretrained Language Models for Graph-to-Text Generation
3
DART: Open-Domain Structured Data Record to Text Generation
4
Bridging the Structural Gap Between Encoding and Decoding for Data-To-Text Generation
5