# Weak Supervision Enhanced Generative Network for Question Generation

## TL;DR

This paper introduces WeGen, a weak supervision-based generative network that enhances question generation by capturing semantic relations between answers and passages, improving quality over existing methods.

## Contribution

The paper proposes a novel weakly supervised framework with a Relation Guider and Multi-Interaction mechanism to better model passage-answer relations for question generation.

## Key findings

- Outperforms baseline models in automatic evaluations
- Achieves higher human evaluation scores
- Effectively captures semantic relations between passage and answer

## Abstract

Automatic question generation according to an answer within the given passage is useful for many applications, such as question answering system, dialogue system, etc. Current neural-based methods mostly take two steps which extract several important sentences based on the candidate answer through manual rules or supervised neural networks and then use an encoder-decoder framework to generate questions about these sentences. These approaches neglect the semantic relations between the answer and the context of the whole passage which is sometimes necessary for answering the question. To address this problem, we propose the Weak Supervision Enhanced Generative Network (WeGen) which automatically discovers relevant features of the passage given the answer span in a weakly supervised manner to improve the quality of generated questions. More specifically, we devise a discriminator, Relation Guider, to capture the relations between the whole passage and the associated answer and then the Multi-Interaction mechanism is deployed to transfer the knowledge dynamically for our question generation system. Experiments show the effectiveness of our method in both automatic evaluations and human evaluations.