Compositional Generalization in Semantic Parsing: Pre-training vs. Specialized Architectures
This addresses the challenge of compositional generalization for semantic parsing tasks, which is incremental as it builds on existing methods to improve performance on benchmarks.
The paper tackled the problem of compositional generalization in semantic parsing by evaluating state-of-the-art techniques and architectures on SCAN and CFQ datasets, showing that masked language model pre-training rivals specialized architectures on primitive splits and achieves significant improvements on complex tasks, establishing a new state of the art on CFQ with an intermediate representation.
While mainstream machine learning methods are known to have limited ability to compositionally generalize, new architectures and techniques continue to be proposed to address this limitation. We investigate state-of-the-art techniques and architectures in order to assess their effectiveness in improving compositional generalization in semantic parsing tasks based on the SCAN and CFQ datasets. We show that masked language model (MLM) pre-training rivals SCAN-inspired architectures on primitive holdout splits. On a more complex compositional task, we show that pre-training leads to significant improvements in performance vs. comparable non-pre-trained models, whereas architectures proposed to encourage compositional generalization on SCAN or in the area of algorithm learning fail to lead to significant improvements. We establish a new state of the art on the CFQ compositional generalization benchmark using MLM pre-training together with an intermediate representation.