Maximum entropy (ME) models have been successfully applied to many natural language problems. In this paper, the authors show how to integrate ME models efficiently within a maximum likelihood training scheme of statistical machine translation models. Specifically, the authors define a set of context-dependent ME lexicon models and show how to perform an efficient training of these ME models within the conventional expectation-maximization (EM) training of statistical translation models. Experimental results are also given in order to demonstrate how these ME models improve the results obtained with the traditional translation models. The results are presented by means of alignment quality comparing the resulting alignments with manually annotated reference alignments.