LoFT: Enhancing Faithfulness and Diversity for Table-to-Text Generation via Logic Form Control
This work addresses the problem of generating accurate and varied text from tables for applications like data summarization, with incremental improvements in combining faithfulness and diversity.
The paper tackles the challenges of generating logically faithful and diverse sentences from tables in Logical Table-to-Text generation, proposing LoFT, which uses logic forms as verifiers and planners, and demonstrates it as the first model to address both issues simultaneously on the LogicNLG dataset.
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.