I was a member of the natural language processing group at Columbia advised by Kathy McKeown and have been involved with the machine learning group and the center for computational learning systems. My current research involves structured prediction approaches to problems such as phrase-based alignment, text-to-text generation and coherence-focused summarization. Since the meaning of any snippet of text is typically spread across multiple words connected via complex latent structures, I'm particularly keen on expressive inference strategies that can simultaneously exploit multiple structural representations of text, e.g., sequences for ordering, trees for syntax, directed acyclic graphs for semantics, etc.

Refereed publications

Dissertations, patents and other publications