I am a member of the Natural Language Processing (NLP) group at Berkeley working with Professor Dan Klein. My research focuses on machine translation (MT), the problem of automatically translating text. While this problem has a long history, a recent shift to statistical techniques has led to substantial improvements in translation quality. Google Translate and Bing Translator both use the approaches I study.

My specific research interest lies in training statistical models of translation. The core of modern translation systems are models that discover the relationships between languages from many examples of human-translated sentences. The group website includes some of our recent published work on MT. With Percy Liang, I maintain the Berkeley Aligner, and unsupervised word alignment system. Adam Pauls and I are also planning to release a syntax-based machine translation pipeline that we have built over the last few years.