Paper: Joint Inference for Heterogeneous Dependency Parsing

ACL ID P13-2019
Title Joint Inference for Heterogeneous Dependency Parsing
Venue Annual Meeting of the Association of Computational Linguistics
Session Short Paper
Year 2013

This paper is concerned with the problem of heterogeneous dependency parsing. In this paper, we present a novel joint infer- ence scheme, which is able to leverage the consensus information between het- erogeneous treebanks in the parsing phase. Different from stacked learning meth- ods (Nivre and McDonald, 2008; Martins et al., 2008), which process the depen- dency parsing in a pipelined way (e.g., a second level uses the first level outputs), in our method, multiple dependency parsing models are coordinated to exchange con- sensus information. We conduct experi- ments on Chinese Dependency Treebank (CDT) and Penn Chinese Treebank (CTB), experimental results show that joint infer- ence can bring significant improvements to all state-of-the-art dependency parsers.