Invited Speakers

Transformation Frameworks for Machine Translation: Strings, Trees, and Graphs

Accurate machine translation (MT) of human languages is a longstanding challenge for computer science. Probabilistic string, tree, and graph automata provide nice formal frameworks for large-scale statistical MT systems. This talk addresses the power of these frameworks and how well they fit observed human translation data.

Alternatives, Discourse Semantics and Discourse Structure

Sentence-level modality and negation give rise to “alternative” events and/or situations that contribute to discourse semantics in interesting ways. But they are not the only linguistic elements that do this. In this talk, I will try to characterise the range of elements that give rise to alternatives and the nature of these events and situations. I will then show how these alternatives are distinct from the coherence relations that provide a low-level of discourse structure.

 
 
speakers.txt · Last modified: 2012/09/13 07:54 by admin
Recent changes RSS feed Creative Commons License Valid XHTML 1.0 Valid CSS Driven by DokuWiki
Drupal Garland Theme for Dokuwiki