Our work involves enriching the Stack-LSTM transition-based AMR parser (Ballesteros and Al-Onaizan, 2017) by augmenting training with Policy Learning and rewarding the Smatch score of sampled graphs. In addition, we also combined several AMR-to-text alignments with an attention mechanism and we supplemented the parser with pre-processed concept identification, named entities and contextualized embeddings. We achieve a highly competitive performance that is comparable to the best published results. We show an in-depth study ablating each of the new components of the parser
@article{arxiv.1905.13370,
title = {Rewarding Smatch: Transition-Based AMR Parsing with Reinforcement Learning},
author = {Tahira Naseem and Abhishek Shah and Hui Wan and Radu Florian and Salim Roukos and Miguel Ballesteros},
journal= {arXiv preprint arXiv:1905.13370},
year = {2019}
}