Dustin’s Path

Publications

Smith, D. and Lieberman, H.
Generating and Interpreting Referring Expressions as Belief State Planning and Plan Recognition.
14th European Workshop on Natural Language Generation, August 2013.

Smith, D. and Lieberman, H.
Using Plan Recognition for Interpreting Referring Expressions.
AAAI Workshop on Plan, Activity, and Intent Recognition, July 2013.

Smith, D and Lieberman, H.
Interpreting Vague and Ambiguous Referring Expressions by Dynamically Binding to Properties of the Context Set.,
January 2013

Smith, D. and Lieberman, H.
Helping users understand and recover from inter- pretation failures in natural language interfaces.
Presented at the Conference on Human Factors in Computing Systems (CHI), May 2012.

Smith, D. and Morgan, B.
IsisWord: An open source common sense simulator for AI researchers.
Presented at the 24th Conference on Artificial Intelligence (AAAI), June 2010.

Smith, D. and Lieberman, H.
The Why UI: Using goal networks to improve user interfaces.
Presented at the 13th International Conference on Intelligent User Interfaces (IUI), February 2010.

Smith, D. and Lieberman, H.
Recognizing and using goals in event management.
Presented at the Conference on Human Factors in Computing Systems (CHI), April 2009.

Dustin graduated from Wake Forest University in 2005, with a Bachelor of Science in Computer Science and a minor in Neuroscience.  Immediately after graduating he was accepted at M.I.T.’s Media Lab, where focusing on Artificial Intelligence.  He received his Masters in 2007, working with advisors, Henry Lebierman and Marvin Minsky. His Media Lab website can be found here.

In 2013, he received his Ph’d at the MIT Media Lab where he developed a system for generating and interpreting referring expressions that include vague and ambiguous modifiers, and explored ways of resolving interpretive uncertainty.  His research goal was to understand the computational-level mechanisms underlying human language production and comprehension, in service of building natural language interfaces so humans and computers can collaborate more effectively.

After receiving his Ph’d, Dr. Smith helped co-found Beansprock with fellow M.I.T. alum Cameron Levy.  On February 4, 2015, Beansprock, a machine learning-based job search platform, launched.  Dustin and Cameron were interviewed in Wired magazine in the article “AI Software That Could Score You the Perfect Job“.

Dustin defending his thesis at M.I.T.

Speer, R., Krishnamurthy, J., Havasi, C., Smith, D., Lieberman, H., Arnold, K.C.
An interface for targeted collection of common sense knowledge using a mixture model.
Proceedings of the 13th International Conference on Intelligent User Interfaces (IUI), February 2009.

Smith, D. and Arnold, K.C.
Learning hierarchical plans by reading simple Eng- lish narratives.
Presented at the Commonsense Workshop at the ACM International Conference on Intelligent User Interfaces (IUI), February 2009.

Smith, D.
EventMinder: A personal calendar assistant that understands events.
Master’s thesis, MIT, September 2007.

Lieberman, H., Smith, D., Teeters, A.
Common Consensus: A Web-based game for collecting common sense goals.
Presented at the Commonsense Workshop at the ACM International Conference on Intelligent User Interfaces (IUI), February 2007.