Deep Semantic Role Labeling: What Works and What’s Next
Luheng Hea b, Kenton Leea b, Mike Lewisc d, Luke Zettlemoyere f
- aSchool of Computer Science & Engineering , Univ. of Washington , Seattle , WA
- bSchool of Computer Science & Engineering, Univ. of Washington, Seattle, WA
- cFacebook AI Research , Menlo Park , CA
- dFacebook AI Research, Menlo Park, CA
- eAllen Institute for Artificial Intelligence , Seattle , WA
- fAllen Institute for Artificial Intelligence, Seattle, WA
Papers open inside SciRah from the original publisher or repository; SciRah never hosts them.
Abstract
We introduce a new deep learning model for semantic role labeling (SRL) that significantly improves the state of the art, along with detailed analyses to reveal its strengths and limitations.We use a deep highway BiLSTM architecture with constrained decoding, while observing a number of recent best practices for initialization and regularization.Our 8-layer ensemble model achieves 83.2 F1 on the CoNLL 2005 test set and 83.4 F1 on CoNLL 2012, roughly a 10% relative error reduction over the previous state of the art.Extensive empirical analysis of these gains show that (1) deep models excel at recovering long-distance dependencies but can still make surprisingly obvious errors, and (2) that there is still room for syntactic parsers to improve these results.
TL;DRA new deep learning model for semantic role labeling (SRL) that significantly improves the state of the art, along with detailed analyses to reveal its strengths and limitations are introduced. (Semantic Scholar)