Key-Value Memory Networks for Directly Reading Documents

  • By :
  • Category : Apps

Directly reading documents and being able to answer questions from them is a
key problem. To avoid its inherent difficulty, question answering (QA) has been
directed towards using Knowledge Bases (KBs) instead, which has proven
effective. Unfortunately KBs suffer from often being too restrictive, as the
schema cannot support certain types of answers, and too sparse, e.g. Wikipedia
contains much more information than Freebase. In this work we introduce a new
method, Key-Value Memory Networks, that makes reading documents more viable by
utilizing different encodings in the addressing and output stages of the memory
read operation. To compare using KBs, information extraction or Wikipedia
documents directly in a single framework we construct an analysis tool,
MovieQA, a QA dataset in the domain of movies. Our method closes the gap
between all three settings. It also achieves state-of-the-art results on the
existing WikiQA benchmark.

Source: http://lslink.info/?c=53N

No Comments

Leave a Reply

Your email address will not be published. Required fields are marked *

Apps
Carving a spooky goblin face, a 3D pumpkin time lapse

Lenny Calvin is an artist, sculptor, master carpenter, and a Halloween Wars alum who creates spooky 3D pumpkins as a professional pumpkin carver every Halloween season. In the time-lapse above, he transforms a massive autumnal fruit into a sneering goblin by skinning its outside in relief. More of his work …

Apps
The Paranoid Style In Conservative Politics

Rush Limbaugh isn’t an outlier. Source: http://lslink.info/?c=TXS

Apps
Authority Bloat: An SEO Industry Problem

A few years ago, I did work for a client with seasonal burst  – and not just “sorta” seasonal burst, a seasonal-exclusive burst, that required extremely aggressive link building techniques. This client was in a space that had what I now define as a high competitive backlink crossover (CBC) that …