Natural Language Processing (NLP) underlies the ability for systems like Siri, and other talk-to-text capabilities to enable computers to derive meaning from what we say. It is a familiar experience however to feel like Siri just doesn't understand what you are trying to say. Luckily, researchers have combined computer science, artificial intelligence, and linguistics to help computers understand not only the literal meaning of a text but also the underlying implicit meaning of a text. Dr. Marie-Catherine de Marneffe, of The Ohio State University, is a linguist that utilizes computational modeling to capture what is conveyed by more than just the literal meaning of words. Her work may therefore aid NLP systems in providing text understanding comparable to what humans achieve and in addition, gain insight into how humans acquire language.

Advances achieved by Dr. de Marneffe's research have the promise of incredible impacts for language technologies. In the current era of ever growing information, having automatic systems that understand the nuances behind speech would benefit information analysts by allowing them to organize and sift through information more quickly and effectively. In fact, Dr. de Marneffe hopes to use such technology to provide a new kind of application that is radically more capable of real understanding that agencies like the NSA could capitalize upon. In addition to applications, Dr. de Marneffe's work has also established closer interactions between researchers in theoretical linguistics and in NLP thereby allowing them to successfully share resources. By leveraging the world of linguistics and computational technology, Dr. de Marneffe provides research that shows how efforts in theoretical linguistics and NLP can complement each other to develop models and to asses them.

Current research includes:

• Targeting Implicit Meaning: Dr. de Marneffe is building automatic systems that target implicit meaning by focusing on different tasks. For example, how can we automatically retrieve that a dialogue such as "Was the movie good? It was excellent." means, "Yes, the movie was good" even though this is not explicitly stated? This research aims at understanding language to allow researchers to suitably organize language systematically.

• Veridicality: Dr. de Marneffe is interested in how we infer meaning from a sentence. For example when one reads something or hears it in the news, how do they decide that the events described are truly reflecting what happens in the world? By better understanding the concept of veridicality, Dr. Marneffe's research may point linguists in the direction of stronger computational tools.

• Language Acquisition: Dr. de Marneffe has combined her NLP work with understanding the foundational issues of how we process language. She has developed empirical, probabilistic models that show how our language production and comprehension rely heavily on keeping track of patterns present in what we hear. This research has principally been interested in looking at how children acquire language.