Language Modeling with Gated Convolutional Networks

Explore this paper's citation graph

Summary

A finite context approach through stacked convolutions, which can be more efficient since they allow parallelization over sequential tokens, is developed and is the first time a non-recurrent approach is competitive with strong recurrent models on these large scale language tasks.

Type
preprint
Published
2016-12-23
Cited by
3,004
References
36
Access
Open access

Keywords

Computer science, Recurrent neural network, Benchmark (surveying), Sentence, Language model

References

Cited by

Related papers