CHIME: Cross-passage Hierarchical Memory Network for Generative Review Question Answering

TL;DR

CHIME is a novel hierarchical memory network that enhances multi-passage generative question answering by integrating cross-passage evidence and answer refinement, leading to more accurate and interpretable responses.

Contribution

The paper introduces CHIME, a new memory-augmented architecture for generative QA that improves answer quality and interpretability over existing models.

Findings

Outperforms state-of-the-art baselines on AmazonQA dataset

Produces more syntactically well-formed answers

Enhances interpretability through memory modules

Abstract

We introduce CHIME, a cross-passage hierarchical memory network for question answering (QA) via text generation. It extends XLNet introducing an auxiliary memory module consisting of two components: the context memory collecting cross-passage evidences, and the answer memory working as a buffer continually refining the generated answers. Empirically, we show the efficacy of the proposed architecture in the multi-passage generative QA, outperforming the state-of-the-art baselines with better syntactically well-formed answers and increased precision in addressing the questions of the AmazonQA review dataset. An additional qualitative analysis revealed the interpretability introduced by the memory module.