Broadening statistical machine translation with comparable corpora and generalized models
Published on Feb 15, 20123003 Views
As we scale statistical machine translation systems to general domain, we face many challenges. This talk outlines two approaches for building better broad-domain systems. First, progress in data-d