Reading Wikipedia to Answer Open-Domain Questions

  • By :
  • Category : Safety

This paper proposes to tackle open- domain question answering using Wikipedia
as the unique knowledge source: the answer to any factoid question is a text
span in a Wikipedia article. This task of machine reading at scale combines the
challenges of document retrieval (finding the relevant articles) with that of
machine comprehension of text (identifying the answer spans from those
articles). Our approach combines a search component based on bigram hashing and
TF-IDF matching with a multi-layer recurrent neural network model trained to
detect answers in Wikipedia paragraphs. Our experiments on multiple existing QA
datasets indicate that (1) both modules are highly competitive with respect to
existing counterparts and (2) multitask learning using distant supervision on
their combination is an effective complete system on this challenging task.

Source: http://bppro.link/?c=QYc

No Comments

Leave a Reply

Safety
Didier Stainier (MPI) 3: Genetic Compensation

Part 1: Vertebrate Organ Development: The Zebrafish Heart: Zebrafish heart development requires the orchestration of cell proliferation, differentiation, and movement. How is this complex process regulated? Part 2: Cardiac Trabeculation: Trabeculae are muscular ridges that form in the heart ventricle and allow it to pump more forcefully.  What controls the …

Safety
French Gift Ideas for Mother’s Day

Buy Clotilde’s latest book, The French Market Cookbook! Mother’s Day is celebrated the second Sunday of May in North America. In France, it is the last Sunday in May. (Unless that falls on Pentecost Sunday, in which case it’s the first Sunday in June. Still with me?) In 2017, Americans …

Safety
How to Generate Revenue With Your Content

Want to make money from your content? Wondering how a loyal audience can create business opportunities? To explore business models that help publishers generate revenue, I interview Joe Pulizzi. More About This Show The Social Media Marketing podcast is an on-demand talk radio show from Social Media Examiner. It’s designed …