Knowledge-based natural answer generation via effective graph learning

doi:10.1016/j.knosys.2025.113288

Knowledge-based natural answer generation via effective graph learning

2025-05-01

0

PRE

AI

DOI：10.1016/j.knosys.2025.113288

Original

Original request for help

Save

Abstract

En 中文

Objectives: Natural Answer Generation (NAG) aims to generate natural and fluent answers to user questions. Existing NAG methods typically employ fixed-hop retrieval to construct knowledge graphs and utilize attention-based networks for answer generation. However, these approaches lack interpretability, struggle to filter out redundant information in the graph, and are computationally intensive. Methods: To address these issues, this paper introduces an innovative approach AdaptQA model. Initially, AdaptQA constructs a knowledge graph from the knowledge base (KB) using an adaptive multi-hop retrieval algorithm. Subsequently, it generates answers through the Graph-based Mamba module (GBM), effectively filtering out redundant information. Finally, the answers are optimized using a pre-trained large language model to enhance their fluency and accuracy. Novelty: The proposed AdaptQA model introduces a new approach to NAG by improving the completeness of the knowledge graph and optimizing question answers. This method overcomes the limitations of existing NAG techniques by reducing the complexity of model inference. Findings: Through extensive experiments on two benchmark datasets, HotpotQA and WikiHop, AdaptQA demonstrates superior performance, significantly outperforming existing NAG methods. Specifically, AdaptQA achieves an accuracy of 94.47% on the HotpotQA dataset and 91.38% on the WikiHop dataset.

Keywords:

Natural answer generation

Adaptive multi-hop retrieval

Graph-based Mamba

Prompt optimization

Knowledge-based natural answer generation via effective graph learning

Abstract

Journal

Organization

Cited Papers