Towards Improving Interpretability of Language Model Generation through a Structured Knowledge Discovery Approach

TL;DR

This paper introduces a structured knowledge discovery approach to improve the interpretability of language model generation by leveraging a hierarchical knowledge representation and a task-agnostic knowledge hunter, enhancing transparency and performance.

Contribution

The paper proposes a novel hierarchical, task-agnostic knowledge hunter using structured knowledge and transformer-based pointer networks to improve interpretability in language generation.

Findings

Outperforms state-of-the-art methods on RotoWireFG and KdConv datasets.

Enhances interpretability by enabling understanding of knowledge selection process.

Achieves higher faithfulness and generalizability in knowledge-enhanced text generation.

Abstract

Knowledge-enhanced text generation aims to enhance the quality of generated text by utilizing internal or external knowledge sources. While language models have demonstrated impressive capabilities in generating coherent and fluent text, the lack of interpretability presents a substantial obstacle. The limited interpretability of generated text significantly impacts its practical usability, particularly in knowledge-enhanced text generation tasks that necessitate reliability and explainability. Existing methods often employ domain-specific knowledge retrievers that are tailored to specific data characteristics, limiting their generalizability to diverse data types and tasks. To overcome this limitation, we directly leverage the two-tier architecture of structured knowledge, consisting of high-level entities and low-level knowledge triples, to design our task-agnostic structured knowledge hunter. Specifically, we employ a local-global interaction scheme for structured knowledge representation learning and a hierarchical transformer-based pointer network as the backbone for selecting relevant knowledge triples and entities. By combining the strong generative ability of language models with the high faithfulness of the knowledge hunter, our model achieves high interpretability, enabling users to comprehend the model output generation process. Furthermore, we empirically demonstrate the effectiveness of our model in both internal knowledge-enhanced table-to-text generation on the RotoWireFG dataset and external knowledge-enhanced dialogue response generation on the KdConv dataset. Our task-agnostic model outperforms state-of-the-art methods and corresponding language models, setting new standards on the benchmark.