Zobrazeno 1 - 10
of 32
pro vyhledávání: '"Nadir Durrani"'
Publikováno v:
ACL/IJCNLP (Findings)
Transfer learning from pre-trained neural language models towards downstream tasks has been a predominant theme in NLP recently. Several researchers have shown that deep NLP models learn non-trivial amount of linguistic knowledge, captured at differe
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::32a12b38da3a8d2d21e92e2590b28dfa
https://doi.org/10.18653/v1/2021.findings-acl.438
https://doi.org/10.18653/v1/2021.findings-acl.438
This paper is a write-up for the tutorial on "Fine-grained Interpretation and Causation Analysis in Deep NLP Models" that we are presenting at NAACL 2021. We present and discuss the research work on interpreting fine-grained components of a model fro
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::b61dfff60642ca63d07d46033b618dcc
Publikováno v:
EMNLP (1)
While a lot of analysis has been carried to demonstrate linguistic knowledge captured by the representations learned within deep NLP models, very little attention has been paid towards individual neurons.We carry outa neuron-level analysis using core
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::f0674b1f4c797d0370c15b635bb8db12
http://arxiv.org/abs/2010.02695
http://arxiv.org/abs/2010.02695
Autor:
Firoj Alam, Fahim Dalvi, Shaden Shaar, Nadir Durrani, Hamdy Mubarak, Alex Nikolov, Giovanni Da San Martino, Ahmed Abdelali, Hassan Sajjad, Kareem Darwish, Preslav Nakov
With the outbreak of the COVID-19 pandemic, people turned to social media to read and to share timely information including statistics, warnings, advice, and inspirational stories. Unfortunately, alongside all this useful information, there was also
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fa02753a553ea1feadcb2863e6e70e24
http://arxiv.org/abs/2007.07996
http://arxiv.org/abs/2007.07996
Publikováno v:
EMNLP (1)
Transformer-based deep NLP models are trained using hundreds of millions of parameters, limiting their applicability in computationally constrained environments. In this paper, we study the cause of these limitations by defining a notion of Redundanc
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::e4fe3830b0351f307b8250b6aec5fef0
https://doi.org/10.18653/v1/2020.emnlp-main.398
https://doi.org/10.18653/v1/2020.emnlp-main.398
Publikováno v:
COLING
Low-resource machine translation suffers from the scarcity of training data and the unavailability of standard evaluation sets. While a number of research efforts target the former, the unavailability of evaluation benchmarks remain a major hindrance
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::529ea4b94e94afd75d1d46e1da99f7a5
https://doi.org/10.18653/v1/2020.coling-main.447
https://doi.org/10.18653/v1/2020.coling-main.447
Transformer-based NLP models are trained using hundreds of millions or even billions of parameters, limiting their applicability in computationally constrained environments. While the number of parameters generally correlates with performance, it is
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::fe8cd2ecf217d5ea8e01eba3f4642185
Publikováno v:
Computer Speech & Language. 45:161-179
Two sets of novel extensions of NNJM model are proposed. The NDAM models that regularizes the loss function with respect to in-domain model, give an improvement of up to +0.4 BLEU points. The NFM models that fuse in- and out-domain NNJM models give a
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::a80420a929f7afad2ff0928bec6488e7
https://doi.org/10.1016/j.csl.2016.12.006
https://doi.org/10.1016/j.csl.2016.12.006
Autor:
Graham Neubig, Yonatan Belinkov, Xian Li, Philipp Koehn, Orhan Firat, Nadir Durrani, Juan Pino, Antonios Anastasopoulos, Hassan Sajjad, Paul Michel
Publikováno v:
WMT (2)
We share the findings of the first shared task on improving robustness of Machine Translation (MT). The task provides a testbed representing challenges facing MT models deployed in the real world, and facilitates new approaches to improve models; rob
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_dedup___::6b53e58659c14a6d8406a771bd74c4eb
Publikováno v:
NAACL-HLT (1)
Recent work has shown that contextualized word representations derived from neural machine translation are a viable alternative to such from simple word predictions tasks. This is because the internal understanding that needs to be built in order to
Externí odkaz:
https://explore.openaire.eu/search/publication?articleId=doi_________::3b9872ccfa8278f3554c7e98a8299b8e
https://doi.org/10.18653/v1/n19-1154
https://doi.org/10.18653/v1/n19-1154