W03-2200 |
directions for improving MT by
|
embedding
|
it in an environment of other
|
W05-0627 |
the largest probability among
|
embedding
|
ones are kept . After predicting
|
W06-0603 |
marker presence and syntactic
|
embedding
|
structure to be strongly associated
|
W06-2205 |
generated from a text corpus by
|
embedding
|
syntactically parsed sentences
|
W07-1022 |
Understanding the structure of the
|
embedding
|
phrase can be an enormously beneficial
|
W10-2802 |
ularities . This latter is employed by
|
embedding
|
prior FrameNet-derived knowledge
|
W10-4104 |
other examples of this section are
|
embedding
|
, while this example is of overlapping
|
W11-1827 |
that is based on the notion of
|
embedding
|
. We apply our methodology to
|
W13-0705 |
what we refer to as its minimum
|
embedding
|
space . The focus here will be
|
W13-3207 |
introduce a new 50-dimensional
|
embedding
|
obtained by spectral clustering
|
W13-3213 |
or not , by classifying its RNN
|
embedding
|
together with those of its siblings
|
W14-1411 |
for the adjectival challenge by
|
embedding
|
the record types defined to deal
|
W14-4002 |
two nonterminal gaps , thereby
|
embedding
|
ITG permutations ( Wu , 1997
|
W14-5201 |
pipelines with other researchers ,
|
embedding
|
NLP pipelines in applications
|
W15-1303 |
relies on the notion of semantic
|
embedding
|
and a fine-grained classification
|
W15-1501 |
on the popular skip-gram word
|
embedding
|
model . The novelty of our approach
|
W15-1504 |
We introduce a new method for
|
embedding
|
word instances and their context
|
W15-1511 |
1992 ) or the ( less well-known )
|
embedding
|
form given by the canonical correlation
|
W15-2608 |
based approach that uses word
|
embedding
|
features to recognize drug names
|
W15-2619 |
rank synonym candidates with word
|
embedding
|
and pseudo-relevance feedback
|