A
Adhiguna Kuncoro
Researcher at Google
Publications - 28
Citations - 2027
Adhiguna Kuncoro is an academic researcher from Google. The author has contributed to research in topics: Language model & Parsing. The author has an hindex of 15, co-authored 25 publications receiving 1751 citations. Previous affiliations of Adhiguna Kuncoro include Carnegie Mellon University & University of Oxford.
Papers
More filters
Posted Content
DyNet: The Dynamic Neural Network Toolkit
Graham Neubig,Chris Dyer,Yoav Goldberg,Austin Matthews,Waleed Ammar,Antonios Anastasopoulos,Miguel Ballesteros,David Chiang,Daniel Clothiaux,Trevor Cohn,Kevin Duh,Manaal Faruqui,Cynthia Gan,Dan Garrette,Yangfeng Ji,Lingpeng Kong,Adhiguna Kuncoro,Gaurav Kumar,Chaitanya Malaviya,Paul Michel,Yusuke Oda,Matthew Richardson,Naomi Saphra,Swabha Swayamdipta,Pengcheng Yin +24 more
TL;DR: DyNet is a toolkit for implementing neural network models based on dynamic declaration of network structure that has an optimized C++ backend and lightweight graph representation and is designed to allow users to implement their models in a way that is idiomatic in their preferred programming language.
Proceedings ArticleDOI
Recurrent Neural Network Grammars
TL;DR: The authors presentada a la 2016 Conference of the North American Chapter of the Association for Computational Linguistics, celebrada a San Diego (CA, EUA) els dies 12 a 17 de juny 2016.
Proceedings ArticleDOI
What Do Recurrent Neural Network Grammars Learn About Syntax
TL;DR: By training grammars without nonterminal labels, it is found that phrasal representations depend minimally on nonterminals, providing support for the endocentricity hypothesis.
Proceedings ArticleDOI
LSTMs Can Learn Syntax-Sensitive Dependencies Well, But Modeling Structure Makes Them Better.
TL;DR: It is found that the mere presence of syntactic information does not improve accuracy, but when model architecture is determined by syntax, number agreement is improved: top-down construction outperforms left-corner and bottom-up variants in capturing non-local structural dependencies.
Posted Content
Recurrent Neural Network Grammars
TL;DR: The authors introduce recurrent neural network grammars, probabilistic models of sentences with explicit phrase structure, which allow application to both parsing and language modeling, and demonstrate that they provide better parsing in English than any single previously published supervised generative model and better language modeling than state-of-the-art sequential RNNs in English and Chinese.