Semi-Supervised QA with Generative Domain-Adaptive Nets

2017/02/07 by Zhilin Yang, Junjie Hu, Yang, Zhilin +5 · 3 citations
Computer Science · #Computation and Language (cs.CL) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Machine Learning (cs.LG) #Natural Language Processing Techniques #Topic Modeling

paper · pdf · doi:10.48550/arxiv.1702.02206

openalex publication_date 2017/02/07 · openalex created_date 2025/10/10 · openalex updated_date 2026/07/28

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.

Citations

• Semi-Supervised Learning with Ladder Networks
• Empirical Evaluation of Gated Recurrent Neural Networks on Sequence Modeling
• Machine Comprehension Using Match-LSTM and Answer Pointer
• NewsQA: A Machine Comprehension Dataset
• Google's Multilingual Neural Machine Translation System: Enabling\n Zero-Shot Translation

Cited by

• Multi-Perspective Context Aggregation for Semi-supervised Cloze-style Reading Comprehension
• Question Generation from SQL Queries Improves Neural Semantic Parsing
• A Study of the Tasks and Models in Machine Reading Comprehension

Related

• End-to-End Synthetic Data Generation for Domain Adaptation of Question Answering Systems
• Domain Adaptation for Question Answering via Question Classification
• Training Generative Question-Answering on Synthetic Data Obtained from an Instruct-tuned Model
• Contrastive Domain Adaptation for Question Answering using Limited Text Corpora
• Learning to Generalize for Cross-domain QA