Rewarding Smatch: Transition-Based AMR Parsing with Reinforcement Learning
This work addresses AMR parsing for NLP applications, but it is incremental as it builds on existing methods with hybrid improvements.
The authors tackled AMR parsing by enhancing a transition-based parser with reinforcement learning to reward Smatch scores, achieving competitive performance comparable to best published results.
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