One-shot Learning for Question-Answering in Gaokao History Challenge

TL;DR

This paper introduces a hybrid neural model for history question-answering in Gaokao exams, combining a cooperative gated neural network with a neural turing machine labeler, achieving significant performance improvements.

Contribution

The work presents a novel hybrid neural architecture that effectively captures semantic relations in complex history questions with limited training data.

Findings

The neural turing machine labeler performs well with small datasets.

The gated mechanism effectively captures semantic representations of lengthy answers.

The proposed model outperforms various neural baselines on multiple metrics.

Abstract

Answering questions from university admission exams (Gaokao in Chinese) is a challenging AI task since it requires effective representation to capture complicated semantic relations between questions and answers. In this work, we propose a hybrid neural model for deep question-answering task from history examinations. Our model employs a cooperative gated neural network to retrieve answers with the assistance of extra labels given by a neural turing machine labeler. Empirical study shows that the labeler works well with only a small training dataset and the gated mechanism is good at fetching the semantic representation of lengthy answers. Experiments on question answering demonstrate the proposed model obtains substantial performance gains over various neural model baselines in terms of multiple evaluation metrics.