Tabula nearly rasa: Probing the Linguistic Knowledge of Character-Level Neural Language Models Trained on Unsegmented Text.

Michael Hahn Marco Baroni

Published in: CoRR (2019)

Keyphrases

language model
linguistic knowledge
document level
language modeling
information retrieval
document retrieval
retrieval model
probabilistic model
text retrieval

query expansion
natural language processing
n gram
natural language
word sense
test collection
word sense disambiguation
text mining
text processing

noun phrases
semantic knowledge
relevance model
sentence level
free text
dialogue system
pseudo relevance feedback
vector space model
cross lingual

low level
query terms
cross language
natural language text
document collections
training set
information extraction
training data
machine learning