Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Unsupervised and weakly supervised approaches for answer selection tasks with scarce annotations

Unsupervised and weakly supervised approaches for answer selection tasks with scarce annotations AbstractAddressing Answer Selection (AS) tasks with complex neural networks typically requires a large amount of annotated data to increase the accuracy of the models. In this work, we are interested in simple models that can potentially give good performance on datasets with no or few annotations. First, we propose new unsupervised baselines that leverage distributed word and sentence representations. Second, we compare the ability of our neural architectures to learn from few annotated examples in a weakly supervised scheme and we demonstrate how these methods can benefit from a pre-training on an external dataset. With an emphasis on results reproducibility, we show that our simple methods can reach or approach state-of-the-art performances on four common AS datasets. http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Open Computer Science de Gruyter

Unsupervised and weakly supervised approaches for answer selection tasks with scarce annotations

Download PDF
9 pages

Loading next page...
 
/lp/de-gruyter/unsupervised-and-weakly-supervised-approaches-for-answer-selection-cPw11ABMe1
Publisher
de Gruyter
Copyright
© 2019 Emmanuel Vallee et al., published by De Gruyter Open
eISSN
2299-1093
DOI
10.1515/comp-2019-0008
Publisher site
See Article on Publisher Site

Abstract

AbstractAddressing Answer Selection (AS) tasks with complex neural networks typically requires a large amount of annotated data to increase the accuracy of the models. In this work, we are interested in simple models that can potentially give good performance on datasets with no or few annotations. First, we propose new unsupervised baselines that leverage distributed word and sentence representations. Second, we compare the ability of our neural architectures to learn from few annotated examples in a weakly supervised scheme and we demonstrate how these methods can benefit from a pre-training on an external dataset. With an emphasis on results reproducibility, we show that our simple methods can reach or approach state-of-the-art performances on four common AS datasets.

Journal

Open Computer Science – de Gruyter

Published: Jan 1, 2019

References

  • SQuAD Questions for Machine Comprehension of arXiv cs
    Rajpurkar
  • Integrating Question Classification and Deep Learning for improved Answer Selection in Proceedings of the th International Conference on Association for Santa Fe New Mexico
    Tayyar Madabushi
  • Attention Is All You Need arXiv cs
    Vaswani
  • This is how we do it : Answer Reranking for Open - domain How Questions with Paragraph Vectors and Minimal Feature in
    Bogdanova
  • Phrase - Indexed Question Answering New Challenge for Scalable Document Comprehension arXiv cs
    Seo
  • Aggregate Model for Matching Sequences arXiv cs
    Wang
  • Representation Learning for Answer Selection with LSTM - Based Importance Weighting th International Conference on Computational Semantics Short papers
    Rücklé
  • What is the Jeopardy Model Quasi - Synchronous Grammar for in Proceedings of the Joint Conference on Empirical Methods in Natural Language Processing and Computational Natural Language
    Wang
  • Sanity Alignment Baseline for Question Answering arXiv cs
    Yadav
  • Distributed of Words and Phrases and their Compositionality in in Information Processing Associates
    Mikolov
  • Cross Domain Regularization for Neural Ranking Models Using Adversarial Learning arXiv cs
    Cohen
  • Two - Step Multi - factor Attention Neural Network for Answer Selection in in
    Zhang
  • The Hungarian method for the assignment problem nav
    Kuhn
  • Automatically Extracting High - Quality Negative Examples for Answer Selection in Question Answering in Proceedings of the th International ACM SIGIR Conference on Research and Development in Information New York
    Zhang
  • on Deep Learning Techniques Applied to Answer Selection in Proceedings of the th International Conference on Association for Santa Fe New Mexico
    Lai
  • Universal Sentence arXiv cs
    Yang
  • Relevance weighting of search terms of the for asi
    Robertson
  • Question Answering through Transfer Learning from Large Fine - grained Supervision Data arXiv cs
    Min
  • Baseline Needs More On Simple Word - Embedding - Based Models and Associated Pooling Mechanisms arXiv cs arXiv cs
    Shen
  • Attention - Based Convolutional Neural Network for Modeling Sentence Pairs arXiv cs
    Yin
  • Improved Matching by Enhancing Mutual in
    Liu
  • If You Can t Beat Them Join Them : Handcrafted Features Complement for Non - Factoid Answer in HLT Association for
    Bogdanova
  • Adam Method for Stochastic Optimization arXiv cs
    Kingma