Smaller Alignment Models for Better Translations: Unsupervised Word Alignment with the l0-norm.

Ashish Vaswani, Liang Huang, David Chiang

Browse the full ACL paper archive.