Sketch and Refine: Towards Faithful and Informative Table-to-Text Generation

TL;DR

This paper introduces SANA, a two-stage table-to-text generation method combining autoregressive skeleton creation with non-autoregressive refinement, significantly improving coverage and faithfulness over previous models.

Contribution

The novel Skeleton-based two-stage approach effectively combines autoregressive and non-autoregressive methods to enhance table-to-text generation quality.

Findings

Outperforms previous state-of-the-art in coverage and faithfulness

Achieves PARENT-T recall of 99.47 on WikiPerson

Significantly improves over existing methods by more than 10 points

Abstract

Table-to-text generation refers to generating a descriptive text from a key-value table. Traditional autoregressive methods, though can generate text with high fluency, suffer from low coverage and poor faithfulness problems. To mitigate these problems, we propose a novel Skeleton-based two-stage method that combines both Autoregressive and Non-Autoregressive generations (SANA). Our approach includes: (1) skeleton generation with an autoregressive pointer network to select key tokens from the source table; (2) edit-based non-autoregressive generation model to produce texts via iterative insertion and deletion operations. By integrating hard constraints from the skeleton, the non-autoregressive model improves the generation's coverage over the source table and thus enhances its faithfulness. We conduct automatic and human evaluations on both WikiPerson and WikiBio datasets. Experimental results demonstrate that our method outperforms the previous state-of-the-art methods in both automatic and human evaluation, especially on coverage and faithfulness. In particular, we achieve PARENT-T recall of 99.47 in WikiPerson, improving over the existing best results by more than 10 points.