About me

I'm a computer scientist working on Machine Learning and its applications to tasks related to modeling human language and vision. Specifically, I am interested in building large-capacity models for both multimodal understanding and multimodal generation. I have done research in areas such as machine learning, natural language understanding and generation, neural & statistical machine translation, syntactic and discourse parsing, question answering, automatic summarization, evaluation methodologies and quality prediction for textual structured output. See my full list of publications .

I am currently a research scientist at Google DeepMind. I have also been a member of Google Research and a member of the Google Translate group in Mountain View, CA. Before joining Google, I was a research scientist at SDL Language Technologies unit after the acquisition of Language Weaver Inc., a Los Angeles--based start-up and the first company to successfully commercialize Statistical Machine Translation.

Prior to that, I received a Ph.D. in Computer Science from USC, while doing research at the Information Sciences Institute (ISI) with Daniel Marcu and Kevin Knight in the Intelligent Systems Division.