Dynamic Few-Shot Learning for Knowledge Graph Question Answering

TL;DR

This paper introduces Dynamic Few-Shot Learning (DFSL), a novel approach that combines in-context learning and semantic similarity to improve knowledge graph question answering, achieving state-of-the-art results across various benchmarks.

Contribution

The paper presents DFSL, a new method that enhances KGQA by effectively leveraging few-shot learning without extensive fine-tuning, improving out-of-domain generalization.

Findings

DFSL achieves state-of-the-art performance on multiple KGQA benchmarks.

DFSL demonstrates strong out-of-domain generalization.

Extensive evaluation confirms DFSL's effectiveness across architectures.

Abstract

Large language models present opportunities for innovative Question Answering over Knowledge Graphs (KGQA). However, they are not inherently designed for query generation. To bridge this gap, solutions have been proposed that rely on fine-tuning or ad-hoc architectures, achieving good results but limited out-of-domain distribution generalization. In this study, we introduce a novel approach called Dynamic Few-Shot Learning (DFSL). DFSL integrates the efficiency of in-context learning and semantic similarity and provides a generally applicable solution for KGQA with state-of-the-art performance. We run an extensive evaluation across multiple benchmark datasets and architecture configurations.