BEGIN:VCALENDAR
VERSION:2.0
PRODID:-//Talks.cam//talks.cam.ac.uk//
X-WR-CALNAME:Talks.cam
BEGIN:VEVENT
SUMMARY:Multilingual NLP via Cross-Lingual Word Embeddings - Ivan Vulic\, 
 LTL\, University of Cambridge
DTSTART:20180427T110000Z
DTEND:20180427T120000Z
UID:TALK103274@talks.cam.ac.uk
CONTACT:Andrew Caines
DESCRIPTION:In the recent past\, NLP as a field has seen tremendous utilit
 y of word embeddings as features in downstream tasks. The fact that these 
 word vectors can be trained on unlabeled monolingual corpora of a language
  makes them an inexpensive resource in NLP. With the increasing use of mon
 olingual word vectors\, there is a need for word vectors that can be used 
 as efficiently across multiple languages as monolingually. Therefore\, lea
 rning bilingual and multilingual word embeddings is currently an important
  research topic. These vectors offer an elegant and language-pair independ
 ent way to represent content across different languages in shared cross-li
 ngual embedding spaces\, and also enable the integration of knowledge from
  external resources (e.g.\, WordNet\, dictionaries) into the embedding spa
 ces. In this talk\, I will briefly discuss the current techniques in cross
 -lingual word embedding learning\, presenting the model typology based on 
 multilingual training data requirements. I will then introduce several ill
 ustrative applications of the induced embedding spaces\, including bilingu
 al dictionary induction\, ad-hoc cross-lingual information retrieval\, and
  cross-lingual transfer for dialogue state tracking.
LOCATION:FW26\, Computer Laboratory
END:VEVENT
END:VCALENDAR