Mixture Content Selection for Diverse Sequence Generation

  title={Mixture Content Selection for Diverse Sequence Generation},
  author={Jaemin Cho and Minjoon Seo and Hannaneh Hajishirzi},
  • Jaemin Cho, Minjoon Seo, Hannaneh Hajishirzi
  • Published 2019
  • Computer Science
  • ArXiv
  • Generating diverse sequences is important in many NLP applications such as question generation or summarization that exhibit semantically one-to-many relationships between source and the target sequences. [...] Key Method The diversification stage uses a mixture of experts to sample different binary masks on the source sequence for diverse content selection. The generation stage uses a standard encoder-decoder model given each selected content from the source sequence.Expand Abstract
    15 Citations

    Figures, Tables, and Topics from this paper.

    Exploring Explainable Selection to Control Abstractive Generation
    Focused Questions and Answer Generation by Key Content Selection
    Text Generation by Learning from Off-Policy Demonstrations
    How Decoding Strategies Affect the Verifiability of Generated Text
    • 10
    • PDF
    Unsupervised Quality Estimation for Neural Machine Translation
    • 6
    • PDF
    CliniQG4QA: Generating Diverse Questions for Domain Adaptation of Clinical Question Answering
    Neural Syntactic Preordering for Controlled Paraphrase Generation
    • 8
    • PDF
    A Mixture of h-1 Heads is Better than h Heads
    • 1
    • PDF


    Get To The Point: Summarization with Pointer-Generator Networks
    • 1,430
    • Highly Influential
    • PDF
    Mixture Models for Diverse Machine Translation: Tricks of the Trade
    • 40
    • Highly Influential
    • PDF
    Guiding Generation for Abstractive Text Summarization Based on Key Information Guide Network
    • 51
    • PDF
    Learn from Your Neighbor: Learning Multi-modal Mappings from Sparse Annotations
    • 4
    • PDF
    Bottom-Up Abstractive Summarization
    • 249
    • Highly Influential
    • PDF
    Diverse Beam Search for Improved Description of Complex Scenes
    • 58
    • Highly Influential
    • PDF
    Sequence to Sequence Mixture Model for Diverse Machine Translation
    • 25
    • PDF
    Analyzing Uncertainty in Neural Machine Translation
    • 83
    • Highly Influential
    • PDF
    Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders
    • 349
    • PDF
    Multiple Choice Learning: Learning to Produce Multiple Structured Outputs
    • 105
    • PDF