Semantic Role Labeling with Iterative Structure Refinement
This work addresses the challenge of capturing non-local dependencies in SRL for natural language processing, representing an incremental improvement over existing methods.
The paper tackled the problem of modeling interactions between argument labeling decisions in Semantic Role Labeling (SRL) by introducing an iterative refinement approach, resulting in state-of-the-art performance on 5 out of 7 CoNLL-2009 languages, including English.
Modern state-of-the-art Semantic Role Labeling (SRL) methods rely on expressive sentence encoders (e.g., multi-layer LSTMs) but tend to model only local (if any) interactions between individual argument labeling decisions. This contrasts with earlier work and also with the intuition that the labels of individual arguments are strongly interdependent. We model interactions between argument labeling decisions through {\it iterative refinement}. Starting with an output produced by a factorized model, we iteratively refine it using a refinement network. Instead of modeling arbitrary interactions among roles and words, we encode prior knowledge about the SRL problem by designing a restricted network architecture capturing non-local interactions. This modeling choice prevents overfitting and results in an effective model, outperforming strong factorized baseline models on all 7 CoNLL-2009 languages, and achieving state-of-the-art results on 5 of them, including English.