13 citations · 17 across the 10 of their papers we have counts for
Showing 2017Show all
2 papers · 1 filter
cs.CL2017
Reconstruction of Word Embeddings from Sub-Word Parameters
Karl Stratos
Pre-trained word embeddings improve the performance of a neural model at the cost of increasing the model size. We propose to benefit from this resource without paying the cost by…
cs.CL2017
A Sub-Character Architecture for Korean Language Processing
Karl Stratos
We introduce a novel sub-character architecture that exploits a unique compositional structure of the Korean language. Our method decomposes each character into a small set of prim…