Grounded Keys-to-Text Generation: Towards Factual Open-Ended Generation

TL;DR

This paper introduces a new grounded keys-to-text generation task, a dedicated dataset, and an automatic metric to improve factual accuracy in open-ended entity description generation, demonstrating significant gains with a specialized model.

Contribution

It proposes a novel grounded keys-to-text task, a new dataset EntDeGen, and an automatic factual correctness metric MAFE, advancing controllability and factual accuracy in open-ended generation.

Findings

MAFE correlates well (60.14) with human judgments.

Rankers improved factual correctness by 15.95% and 34.51%.

Combining keys and groundings benefits generation quality.

Abstract

Large pre-trained language models have recently enabled open-ended generation frameworks (e.g., prompt-to-text NLG) to tackle a variety of tasks going beyond the traditional data-to-text generation. While this framework is more general, it is under-specified and often leads to a lack of controllability restricting their real-world usage. We propose a new grounded keys-to-text generation task: the task is to generate a factual description about an entity given a set of guiding keys, and grounding passages. To address this task, we introduce a new dataset, called EntDeGen. Inspired by recent QA-based evaluation measures, we propose an automatic metric, MAFE, for factual correctness of generated descriptions. Our EntDescriptor model is equipped with strong rankers to fetch helpful passages and generate entity descriptions. Experimental result shows a good correlation (60.14) between our proposed metric and human judgments of factuality. Our rankers significantly improved the factual correctness of generated descriptions (15.95% and 34.51% relative gains in recall and precision). Finally, our ablation study highlights the benefit of combining keys and groundings.