In the early days of email, widely-used conventions for indicating quoted reply content and email signatures made it easy to segment email messages into their functional parts. To...
We present SpeedBoost, a natural extension of functional gradient descent, for learning anytime predictors, which automatically trade computation time for predictive accuracy by s...
Parallel data in the domain of interest is the key resource when training a statistical machine translation (SMT) system for a specific purpose. Since ad-hoc manual translation c...
Prasanth Kolachina, Nicola Cancedda, Marc Dymetman...
Overlap in markup occurs where some markup structures do not nest, such as where the sentence and phrase boundaries of a poem and the metrical line structure describe different hi...
Fully automatic machine translation cannot produce high quality translation; Dialog-Based Machine Translation (DBMT) is the only way to provide authors with a means of translating...