Humpty Dumpty: Controlling Word Meanings via Corpus Poisoning

Schuster Tal
Schuster Tal
Meri Yoav
Meri Yoav

IEEE Symposium on Security and Privacy, pp. 1295-1313, 2020.

Cited by: 0|Bibtex|Views20|DOI:https://doi.org/10.1109/SP40000.2020.00115
EI
Other Links: arxiv.org|academic.microsoft.com|dblp.uni-trier.de

Abstract:

Word embeddings, i.e., low-dimensional vector representations such as GloVe and SGNS, encode word \"meaning\" in the sense that distances between words’ vectors correspond to their semantic proximity. This enables transfer learning of semantics for a variety of natural language processing tasks.Word embeddings are typically trained on lar...More

Code:

Data:

Full Text
Your rating :
0

 

Tags
Comments