ArcaneQA: Dynamic Program Induction and Contextualized Encoding for Knowledge Base Question Answering

TL;DR

ArcaneQA introduces a generation-based approach with dynamic program induction and contextualized encoding to improve knowledge base question answering, effectively handling large search spaces and schema ambiguities.

Contribution

It presents a novel unified framework combining dynamic program induction and contextualized encoding for KBQA, outperforming traditional ranking-based models.

Findings

Achieves state-of-the-art results on multiple KBQA datasets.

Demonstrates improved efficiency over ranking-based models.

Effectively handles complex queries and schema ambiguities.

Abstract

Question answering on knowledge bases (KBQA) poses a unique challenge for semantic parsing research due to two intertwined challenges: large search space and ambiguities in schema linking. Conventional ranking-based KBQA models, which rely on a candidate enumeration step to reduce the search space, struggle with flexibility in predicting complicated queries and have impractical running time. In this paper, we present ArcaneQA, a novel generation-based model that addresses both the large search space and the schema linking challenges in a unified framework with two mutually boosting ingredients: dynamic program induction for tackling the large search space and dynamic contextualized encoding for schema linking. Experimental results on multiple popular KBQA datasets demonstrate the highly competitive performance of ArcaneQA in both effectiveness and efficiency.