Abstract
This paper describes work on using Minimal Recursion Semantics (MRS) representations for the task of recognising textual entailment. I use entailment data from a SemEval-2010 shared task to develop and evaluate an entailment recognition heuristic. I compare my results to the shared task winner, and discuss differences in approaches. Finally, I run my system with multiple MRS representations per sentence, and show that this improves the recognition results for positive entailment sentence pairs.
Proceedings of the Student Research Workshop at the 14th Conference of the European Chapter of the Association for Computational Linguistics, 2014, Pages 76-84.