Beyond Log-Linear Models: Boosted Minimum Error Rate Training for N-best Re-ranking.

Kevin Duh, Katrin Kirchhoff

Browse the full ACL paper archive.