Paper: CNGL-CORE: Referential Translation Machines for Measuring Semantic Similarity

ACL ID S13-1034
Title CNGL-CORE: Referential Translation Machines for Measuring Semantic Similarity
Venue Joint Conference on Lexical and Computational Semantics
Session
Year 2013
Authors

We invent referential translation machines (RTMs), a computational model for identify- ing the translation acts between any two data sets with respect to a reference corpus selected in the same domain, which can be used for judging the semantic similarity between text. RTMs make quality and semantic similarity judgments possible by using retrieved rele- vant training data as interpretants for reach- ing shared semantics. An MTPP (machine translation performance predictor) model de- rives features measuring the closeness of the test sentences to the training data, the diffi- culty of translating them, and the presence of acts of translation involved. We view seman- tic similarity as paraphrasing between any two given texts. Each view is modeled by an RTM model, giving us a new perspective o...