Towards Faithfulness in Open Domain Table-to-text Generation from an Entity-centric View

TL;DR

This paper introduces entity-centric metrics to evaluate faithfulness in open domain table-to-text generation, analyzes the impact of data quality, and proposes methods to improve generation fidelity through entity information and instance selection.

Contribution

It presents novel entity-centric metrics for faithfulness evaluation and proposes two methods—augmented training and instance selection—to enhance generation fidelity.

Findings

Metrics align well with human judgments.

Data quality correlates with generation faithfulness.

Proposed methods improve fidelity in various settings.

Abstract

In open domain table-to-text generation, we notice that the unfaithful generation usually contains hallucinated content which can not be aligned to any input table record. We thus try to evaluate the generation faithfulness with two entity-centric metrics: table record coverage and the ratio of hallucinated entities in text, both of which are shown to have strong agreement with human judgements. Then based on these metrics, we quantitatively analyze the correlation between training data quality and generation fidelity which indicates the potential usage of entity information in faithful generation. Motivated by these findings, we propose two methods for faithful generation: 1) augmented training by incorporating the auxiliary entity information, including both an augmented plan-based model and an unsupervised model and 2) training instance selection based on faithfulness ranking. We show these approaches improve generation fidelity in both full dataset setting and few shot learning settings by both automatic and human evaluations.