Sampling-based approximations to minimum Bayes risk decoding for neural machine translation
In NMT we search for the mode of the model distribution to form predictions. The mode and
other high-probability translations found by beam search have been shown to often be …
other high-probability translations found by beam search have been shown to often be …
[图书][B] Quality estimation for machine translation
Many applications within natural language processing involve performing text-to-text
transformations, ie, given a text in natural language as input, systems are required to …
transformations, ie, given a text in natural language as input, systems are required to …
[PDF][PDF] Lattice Minimum Bayes-Risk decoding for statistical machine translation
Abstract We present Minimum Bayes-Risk (MBR) decoding over translation lattices that
compactly encode a huge number of translation hypotheses. We describe conditions on the …
compactly encode a huge number of translation hypotheses. We describe conditions on the …
Sampling-Based Approximations to Minimum Bayes Risk Decoding for Neural Machine Translation
In NMT we search for the mode of the model distribution to form predictions. The mode and
other high-probability translations found by beam search have been shown to often be …
other high-probability translations found by beam search have been shown to often be …
[PDF][PDF] Unsupervised transcription of historical documents
T Berg-Kirkpatrick, G Durrett… - Proceedings of the 51st …, 2013 - aclanthology.org
We present a generative probabilistic model, inspired by historical printing processes, for
transcribing images of documents from the printing press era. By jointly modeling the text of …
transcribing images of documents from the printing press era. By jointly modeling the text of …
[图书][B] Coarse-to-fine natural language processing
S Petrov - 2011 - books.google.com
The impact of computer systems that can understand natural language will be tremendous.
To develop this capability we need to be able to automatically and efficiently analyze large …
To develop this capability we need to be able to automatically and efficiently analyze large …
[PDF][PDF] Efficient minimum error rate training and minimum bayes-risk decoding for translation hypergraphs and lattices
S Kumar, W Macherey, C Dyer… - … of the ACL and the 4th …, 2009 - aclanthology.org
Abstract Minimum Error Rate Training (MERT) and Minimum Bayes-Risk (MBR) decoding
are used in most current state-of-theart Statistical Machine Translation (SMT) systems. The …
are used in most current state-of-theart Statistical Machine Translation (SMT) systems. The …
Bagging and boosting statistical machine translation systems
In this article we address the issue of generating diversified translation systems from a single
Statistical Machine Translation (SMT) engine for system combination. Unlike traditional …
Statistical Machine Translation (SMT) engine for system combination. Unlike traditional …
Post-ordering by parsing with ITG for Japanese-English statistical machine translation
Word reordering is a difficult task for translation between languages with widely different
word orders, such as Japanese and English. A previously proposed post-ordering method …
word orders, such as Japanese and English. A previously proposed post-ordering method …
[PDF][PDF] Quadratic-time dependency parsing for machine translation
M Galley, CD Manning - Proceedings of the Joint Conference of …, 2009 - aclanthology.org
Efficiency is a prime concern in syntactic MT decoding, yet significant developments in
statistical parsing with respect to asymptotic efficiency haven't yet been explored in MT …
statistical parsing with respect to asymptotic efficiency haven't yet been explored in MT …