Zobrazeno 1 - 4
of 4
pro vyhledávání: '"Yury Zemlyanskiy"'
Publikováno v:
NAACL-HLT
Knowledge-intensive tasks such as question answering often require assimilating information from different sections of large inputs such as books or article collections. We propose ReadTwice, a simple and effective technique that combines several str
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f0eac85489298b7d916fc6376bffcc17
http://arxiv.org/abs/2105.04241
http://arxiv.org/abs/2105.04241
Autor:
Ilya Eckstein, Yury Zemlyanskiy, Sudeep Gandhe, Anirudh Ravula, Juraj Gottweis, Bhargav Kanagal, Ruining He, Fei Sha
Publikováno v:
EACL
This paper explores learning rich self-supervised entity representations from large amounts of the associated text. Once pre-trained, these models become applicable to multiple entity-centric tasks such as ranked retrieval, knowledge base completion,
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::094c10a9de68867679c891bef703316a
https://doi.org/10.18653/v1/2021.eacl-main.217
https://doi.org/10.18653/v1/2021.eacl-main.217
Publikováno v:
ACL (1)
There exist few text-specific methods for unsupervised anomaly detection, and for those that do exist, none utilize pre-trained models for distributed vector representations of words. In this paper we introduce a new anomaly detection method—Contex
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::238a3730313e447627f2a4834f9b1884
https://doi.org/10.18653/v1/p19-1398
https://doi.org/10.18653/v1/p19-1398
Autor:
Fei Sha, Yury Zemlyanskiy
Publikováno v:
CoNLL
There have been several attempts to define a plausible motivation for a chit-chat dialogue agent that can lead to engaging conversations. In this work, we explore a new direction where the agent specifically focuses on discovering information about i
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::94d8c485d06eee69e95257d667cec400
https://doi.org/10.18653/v1/k18-1053
https://doi.org/10.18653/v1/k18-1053