Daniel Bauer

Daniel Bauer, PhD Candidate
Columbia University

Department of Computer Science
1214 Amsterdam Avenue
450 Computer Science Building
Mail Code 0401
New York, NY 10027

Office: 7LW3 Shapiro CEPSR
Phone: +1 (212) 870-1274
E-mail: last name at cs dot institution .edu

Currently, I am a Computer Science PhD Candidate at Columbia University in the City of New York where I am affiliated with the Natural Language Processing Group and the Center for Computational Learning Systems (CCLS). My PhD advisor is Owen Rambow. My main research focus is on the interaction between natural language syntax and semantics.

I am a co-founder of WordsEye, a startup company that takes text-to-scene generation, the automatic genertion of 3D scenes from natural language descriptions, to the social media market.

Previously, I have obtained a MSc degree in Language Science and Technology from Saarland University, Germany and a BSc degree in Cognitive Science from the University of Osnabrück, Germany.


My PhD research is on semantic parsing of natural language into graph-based meaning representations, using formal grammar methods. In my thesis, I develop algorithms that can automatically learn semantic parsers from annotated text corpora, as well as efficient statistical parsing algorithms to translate strings into graphs. I apply these techniques to a number of different data sets and meaning representations, including Abstract Meaning Representation and WordsEye Scene Graphs.

My interests also include computational and lexical semantics, especially figuring out how language understanding works in context of linguistic discourse and the real world.

I am one of the authors of the Bolinas toolkit for Synchronous Hyperedge Replacement Grammar.

My masters thesis was on statistical generation with tree adjoining grammars.


I have been the instructor for several undergraduate level CS classes at Columbia.

Fall 2015: COMS W3134, Section I - Data Structures in Java.

Spring 2015: COMS W3134 - Data Structures in Java.

Fall 2014: Two sections COMS W3101 - Programming Languages: Python (1st half of fall), Scala (2nd half of fall).

Spring 2012: COMS W3101-3 - Programming Languages: Python.

TA-ing: Fall 2011: COMS W4705 - Introduction to Natural Language Processing (taught by Michael Collins).

PhD Milestones

On Mar 11 2013 I passed my PhD Candidacy Exam on Semantic Parsing.


I was an issue editor for the ACM XRDS Fall 2014 Issue on Natural Language.

I no longer organize the NLP seminar series ( NLP meetings) at Columbia. If you are interested in giving a talk, please get in touch with Chris Kedzie.