Sequence-to-sequence AMR Parsing with Ancestor Information

AMR parsing is the task of mapping a sentence to an AMR semantic graph automatically. The difficulty comes from generating the complex graph structure. The previous state-of-the-art method translates the AMR graph into a sequence, then directly fine-tunes a pretrained sequence-to-sequence Transformer model (BART). However, purely treating the graph as a sequence does not take advantage of structural information about the graph. In this paper, we design several strategies to add the important ancestor information into the Transformer Decoder. Our experiments1 show that we can improve the performance for both the AMR 2.0 and AMR 3.0 dataset and achieve new state-of-the-art results.