HackerPapers

arXiv:1611.01604cs.CL

Dynamic Coattention Networks For Question Answering

Caiming Xiong, Victor Zhong, Richard Socher

Abstract

Several deep learning models have been proposed for question answering. However, due to their single-pass nature, they have no way to recover from local maxima corresponding to incorrect answers. To address this problem, we introduce the Dynamic Coattention Network (DCN) for question answering. The DCN first fuses co-dependent representations of the question and the document in order to focus on relevant parts of both. Then a dynamic pointing decoder iterates over potential answer spans. This iterative procedure enables the model to recover from initial local maxima corresponding to incorrect answers. On the Stanford question answering dataset, a single DCN model improves the previous state of the art from 71.0% F1 to 75.9%, while a DCN ensemble obtains 80.4% F1.

arXiv comments
14 pages, 7 figures, International Conference on Learning Representations 2017
Journal reference
DOI
Subjects
cs.CLComputation and Language cs.AIArtificial Intelligence

On Hacker News

Linked by 2 identifiable commenters in 2 comments across 2 threads, from Apr 2018 to Sep 2022.

Comments linking this paper, per year.

Discussed in comments2 threads

  1. Ask HN: What is the AI that Ray Kurzweil talks about in Lex Friedman? 1 comment

    • I suspect it’s this model or something similar: https://arxiv.org/abs/1611.01604 Question Answering is a rich area of research in NLP - this is a good place to start: https://paperswithcode.com/task/question-answering

      alexcnwyView on HNParent context

  2. Attention and Augmented Recurrent Neural Networks 1 comment

    • The more useful part about neural net in NLP is that you can train complexe end to end models that end up performing better at their task than anything else composite. An example of this is neural translation with the Attention is all you need paper…

      make3View on HNParent context

Submitted to Hacker News

Connected through shared HN threads, not inferred similarity or a recommendation.