Weak Supervision Enhanced Generative Network for Question Generation
This work addresses question generation for applications like question answering and dialogue systems, presenting an incremental improvement over prior neural-based methods.
The paper tackles the problem of generating questions from a passage and answer by addressing the neglect of semantic relations between the answer and the entire passage in existing methods, proposing the Weak Supervision Enhanced Generative Network (WeGen) that improves question quality through a weakly supervised approach.
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.