Online Back-Parsing for AMR-to-Text Generation
This addresses the challenge of generating accurate text from AMR graphs for natural language processing applications, representing an incremental improvement.
The paper tackles the problem of AMR-to-text generation by proposing a decoder that back-predicts projected AMR graphs during text generation, resulting in outputs that better preserve input meaning, with experiments showing superiority over the previous state-of-the-art graph Transformer system on two benchmarks.
AMR-to-text generation aims to recover a text containing the same meaning as an input AMR graph. Current research develops increasingly powerful graph encoders to better represent AMR graphs, with decoders based on standard language modeling being used to generate outputs. We propose a decoder that back predicts projected AMR graphs on the target sentence during text generation. As the result, our outputs can better preserve the input meaning than standard decoders. Experiments on two AMR benchmarks show the superiority of our model over the previous state-of-the-art system based on graph Transformer.