LoFT: Enhancing Faithfulness and Diversity for Table-to-Text Generation via Logic Form Control

TL;DR

LoFT introduces a novel approach using logic forms to improve factual accuracy and diversity in table-to-text generation, effectively addressing two key challenges in the field.

Contribution

The paper presents LoFT, a model that employs logic forms as fact verifiers and content planners, simultaneously enhancing faithfulness and diversity in table-to-text generation.

Findings

Outperforms previous models on LogicNLG dataset

Addresses unfaithfulness and lack of diversity simultaneously

Code is publicly available for reproducibility

Abstract

Logical Table-to-Text (LT2T) generation is tasked with generating logically faithful sentences from tables. There currently exists two challenges in the field: 1) Faithfulness: how to generate sentences that are factually correct given the table content; 2) Diversity: how to generate multiple sentences that offer different perspectives on the table. This work proposes LoFT, which utilizes logic forms as fact verifiers and content planners to control LT2T generation. Experimental results on the LogicNLG dataset demonstrate that LoFT is the first model that addresses unfaithfulness and lack of diversity issues simultaneously. Our code is publicly available at https://github.com/Yale-LILY/LoFT.