Fetching the paper…

Rewarding Smatch: Transition-Based AMR Parsing with Reinforcement Learning · Around