Joshua is an open-source statistical machine translation decoder for hierarchical and syntax-based machine translation, written in Java. It is developed as a research collaboration between the Human Language Technology Center of Excellence at Johns Hopkins University and Chris Callison-Burch's research group at the University of Pennsylvania.
The latest version of Joshua is version 5.0. Its changes are described in the paper
Joshua 5.0: Sparser, better, faster, server BibTeX
Matt Post, Juri Ganitkevitch, Luke Orland, Jonathan Weese, Yuan Cao, and Chris Callison-Burch
WMT 2013. Sofia, Bulgaria. August.
You might also be interested in older papers describing Joshua 1.0, the Joshua demo paper, 2.0, 3.0, and 4.0.
Technical Support
Support is provided in two forms: Technical support for end-users of Joshua, and Developer support for contributors.
Contributors
Joshua was originally ported from David Chiang's Python implementation of Hiero by Zhifei Li. Since then, there have been a number of contributors to the project (listed here in alphabetical order).
- Aaron Phillips (Carnegie Mellon)
- Adam Pauls (Berkeley)
- Chris Callison-Burch (Penn)
- Chris Dyer (Carnegie Mellon)
- Colin Bannard (UT Austin)
- Jonathan Weese (Johns Hopkins)
- Josh Schroeder (Edinburgh)
- Juri Ganitkevitch (Johns Hopkins)
- Kenneth Heafield (Carnegie Mellon)
- Lane Schwartz (Minnesota)
- Luke Orland (Johns Hopkins)
- Matt Post (Johns Hopkins)
- Omar F. Zaidan (Johns Hopkins)
- Wren Ng Thornton (Indiana University)
- Yuan Cao (Johns Hopkins)
Acknowledgments
We gratefully acknowledge support provided by the following institutions.
