Semi-Supervised QA with Generative Domain-Adaptive Nets

TL;DR

This paper introduces a semi-supervised question answering framework that leverages unlabeled text through generative models and domain adaptation techniques, significantly improving QA performance.

Contribution

It presents a novel generative domain-adaptive training framework for semi-supervised QA, combining question generation with reinforcement learning-based domain adaptation.

Findings

Substantial performance improvements using unlabeled data

Effective domain adaptation reduces data distribution mismatch

Generative models enhance question answering accuracy

Abstract

We study the problem of semi-supervised question answering----utilizing unlabeled text to boost the performance of question answering models. We propose a novel training framework, the Generative Domain-Adaptive Nets. In this framework, we train a generative model to generate questions based on the unlabeled text, and combine model-generated questions with human-generated questions for training question answering models. We develop novel domain adaptation algorithms, based on reinforcement learning, to alleviate the discrepancy between the model-generated data distribution and the human-generated data distribution. Experiments show that our proposed framework obtains substantial improvement from unlabeled text.