Data-to-Text Generation with Iterative Text Editing

TL;DR

This paper introduces an iterative text editing approach for data-to-text generation that enhances semantic accuracy and fluency by combining neural models with filtering and reranking techniques.

Contribution

It proposes a novel iterative editing framework leveraging pre-trained models for improved data-to-text generation and explores zero-shot domain adaptation capabilities.

Findings

Achieves higher semantic accuracy and fluency in generated texts.

Effective on major datasets like WebNLG and E2E.

Enables zero-shot domain adaptation for data-to-text tasks.

Abstract

We present a novel approach to data-to-text generation based on iterative text editing. Our approach maximizes the completeness and semantic accuracy of the output text while leveraging the abilities of recent pre-trained models for text editing (LaserTagger) and language modeling (GPT-2) to improve the text fluency. To this end, we first transform data items to text using trivial templates, and then we iteratively improve the resulting text by a neural model trained for the sentence fusion task. The output of the model is filtered by a simple heuristic and reranked with an off-the-shelf pre-trained language model. We evaluate our approach on two major data-to-text datasets (WebNLG, Cleaned E2E) and analyze its caveats and benefits. Furthermore, we show that our formulation of data-to-text generation opens up the possibility for zero-shot domain adaptation using a general-domain dataset for sentence fusion.