Monday, May 9, 2016 - 13:30
Room: 

The Blocks World Redux

This talk will provide an overview of a new DARPA program focused on Communicating with Computers, CwC.  One of the applications in CwC is a Blocks World domain, where the actions are very simple and concrete, such as “Add a block to the tower.”  However, even in this restricted world, getting the appropriate contextual interpretation of a sentence can be challenging.  The talk will review the progress we have made so far on achieving the goal of contextual interpretation.  This requires the bringing to bear of many resources, including James Allen’s TRIPS ontology and parser, Jerry Hobbs’s axiomatization of many object and action definitions, James Pustejovsky’s Generative Lexicon (GL), and VerbNet (VN).  A main focus of the talk will be the ways in which we are modifying VerbNet to more closely align with Generative Lexicon Event Structure representations, to produce GL—VN.  Working with Julia Hockenmaier, we are also synchronizing these VerbNet with the lexical entries of a Combinatory Categorial Grammar (CCG) semantic parser so that it produces the appropriate sentence representations, including implicit arguments, for the Blocks World domain.  This CCG/VN semantic parser approach will also be described. The talk will conclude with both short term and long term goals for our collaborations on CwC, with respect to  both GL-VN and CCG-VN.

CV: 
Martha Palmer is a Professor at the University of Colorado with joint appointments in Linguistics and Computer Science and is an Institute of Cognitive Science Faculty Fellow. Her Ph.D. is in Artificial Intelligence from the University of Edinburgh. She is an Association of Computational Linguistics (ACL) Fellow, and has won an Outstanding Graduate Advisor 2014 Award, a Boulder Faculty Assembly 2010 Research Award and was the Director of the 2011 Linguistics Institute in Boulder, CO. Her research is focused on capturing elements of the meanings of words that can comprise automatic representations of complex sentences and documents. Supervised machine learning techniques rely on vast amounts of annotated training data so she and her students are engaged in providing data with semantic annotation for English, Chinese, Arabic, Hindi, and Urdu, funded by DARPA and NSF, both manually and automatically. A more recent focus is the application of these methods to biomedical journal articles and clinical notes, funded by NIH. She co-edits LiLT, Linguistic Issues in Language Technology, and has been a co-editor of the Journal of Natural Language Engineering and on the CLJ Editorial Board. She is a past President of ACL, and past Chair of SIGLEX and of SIGHAN.