SUBS: Subtree Substitution for Compositional Semantic Parsing. (arXiv:2205.01538v1 [cs.CL])

Although sequence-to-sequence models often achieve good performance in
semantic parsing for i.i.d. data, their performance is still inferior in
compositional generalization. Several data augmentation methods have been
proposed to alleviate this problem. However, prior work only leveraged
superficial grammar or rules for data augmentation, which resulted in limited
improvement. We propose to use subtree substitution for compositional data
augmentation, where we consider subtrees with similar semantic functions as
exchangeable. Our experiments showed that such augmented data led to
significantly better …

arxiv parsing semantic