AMR-to-text generation as a Traveling Salesman Problem

Linfeng Song1, Yue Zhang2, Xiaochang Peng1, Zhiguo Wang3, Daniel Gildea1
1University of Rochester, 2Singapore University of Technology and Design, 3IBM Watson Research Center


Abstract

The task of AMR-to-text generation is to gen- erate grammatical text that sustains the seman- tic meaning for a given AMR graph. We at- tack the task by first partitioning the AMR graph into smaller fragments, and then gener- ating the translation for each fragment, before finally deciding the order by solving an asym- metric generalized traveling salesman prob- lem (AGTSP). A Maximum Entropy classifier is trained to estimate the traveling costs, and a TSP solver is used to find the optimized solu- tion. The final model reports a BLEU score of 22.44 on the SemEval-2016 Task8 dataset.