Given a number of machine translations of a source segment, the goal of system combination is to produce a new translation that has better quality than all of them. This paper describes a number of improvements that were recently added to the JHU system combination scheme: (i) A hypothesis ranking technique which orders the system outputs, on a per-segment basis, according to predicted translation quality, thus improving a subsequent incremental combination step. (ii) A two-pass combination procedure, which first produces several combination outputs with the given translations, and then performs one more combination step with these new outputs. Results from the NIST MT09 informal system combination evaluation on Arabic-to-English and Urdu-to-English1 show that both approaches offer significant BLEU and TER gains over a baseline JHU combination scheme.

We present minimum Bayes-risk system combination, a method that integrates consensus decoding and system combination into a unified multi-system minimum Bayes-risk (MBR) technique. Unlike other MBR methods that re-rank translations of a single SMT system, MBR system combination uses the MBR decision rule and a linear combination of the component systems ’ probability distributions to search for the minimum risk translation among all the finite-length strings over the output vocabulary. We introduce expected BLEU, an approximation to the BLEU score that allows to efficiently apply MBR in these conditions. MBR system combination is a general method that is independent of specific SMT models, enabling us to combine systems with heterogeneous structure. Experiments show that our approach bring significant improvements to single-system-based MBR decoding and achieves comparable results to different state-of-the-art system combination methods. 1

This paper describes the JHU system combination scheme used in WMT-11. The JHU system combination is based on confusion network alignment, and inherited the framework developed by (Karakos et al., 2008). We improved our core system combination algorithm by making use of TER-plus, which was originally designed for string alignment, for alignment of confusion networks. Experimental results on French-English, German-English, Czech-English and Spanish-English combination tasks show significant improvements on BLEU and TER by up to 2 points on average, compared to the best individual system output, and improvements compared with the results produced by ITG which we used in WMT-10. 1