Cross-Lingual Training for Automatic Question Generation
Vishwajeet Kumar IITB-Monash Research Academy, Mumbai, India IIT Bombay, Mumbai, India Nitish Joshi IIT Bombay, Mumbai, India Arijit Mukherjee IIT Bombay, Mumbai, India Ganesh Ramakrishnan IIT Bombay, Mumbai, India Preethi Jyothi IIT Bombay, Mumbai, India
Abstract
Automatic question generation (QG) is a challenging problem in natural language understanding. QG systems are typically built assuming access to a large number of training instances where each instance is a question and its corresponding answer. For a new language, such training instances are hard to obtain making the QG problem even more challenging. Using this as our motivation, we study the reuse of an available large QG dataset in a secondary language (e.g. English) to learn a QG model for a primary language (e.g. Hindi) of interest. For the primary language, we assume access to a large amount of monolingual text but only a small QG dataset. We propose a cross-lingual QG model which uses the following training regime: (i) Unsupervised pretraining of language models in both primary and secondary languages and (ii) joint supervised training for QG in both languages. We demonstrate the efficacy of our proposed approach using two different primary languages, Hindi and Chinese. We also create and release a new question answering dataset for Hindi consisting of 6555 sentences.
中文速览
为低资源语言自动生成问题（Question Generation, QG）一直很困难，因为这类语言几乎没有足够的带标注训练数据。研究团队提出了一种跨语言QG模型：先用两种语言的单语语料做无监督预训练（去噪自编码器＋回译），让模型学会语言间的共享表示，再用英语的大规模QG数据集与目标语言（印地语或中文）的少量QG数据联合有监督微调，从而把英语的资源"迁移"给资源匮乏的语言。实验表明，这一方法在印地语和中文上的问题生成质量均显著优于仅使用目标语言小数据训练的基线模型。此外，团队还发布了首个印地语问答数据集HiQuAD（6555个句子-问题对），为后续研究提供了重要的基准资源。
原文 arXiv:1906.02525;中英对照 + 大白话阅读 https://aha.fim.ai/paper/1906.02525v1