Syntactic Scaffolds for Semantic Structures
This addresses the challenge of efficiently leveraging syntax for semantic tasks in NLP, but it is incremental as it builds on existing multitask learning approaches.
The paper tackles the problem of incorporating syntactic information into semantic tasks without runtime overhead by using a multitask objective during training, achieving competitive performance on PropBank semantics, frame semantics, and coreference resolution.
We introduce the syntactic scaffold, an approach to incorporating syntactic information into semantic tasks. Syntactic scaffolds avoid expensive syntactic processing at runtime, only making use of a treebank during training, through a multitask objective. We improve over strong baselines on PropBank semantics, frame semantics, and coreference resolution, achieving competitive performance on all three tasks.