Desrl: A linear-time semantic role labeling system

This paper describes the DeSRL sys-tem, a joined effort of Yahoo! Research Barcelona and Universita ̀ di Pisa for the CoNLL-2008 Shared Task (Surdeanu et al., 2008). The system is characterized by an efficient pipeline of linear complexity components, each carrying out a different sub-task. Classifier errors and ambigui-ties are addressed with several strategies: revision models, voting, and reranking. The system participated in the closed chal-lenge ranking third in the complete prob-lem evaluation with the following scores: 82.06 labeled macro F1 for the overall task, 86.6 labeled attachment for syntactic de-pendencies, and 77.5 labeled F1 for se-mantic dependencies. 1 System description DeSRL is implemented as a sequence of compo-nents of linear complexity relative to the sentence length. We decompose the problem into three sub-tasks: parsing, predicate identification and clas-sification (PIC), and argument identification and classification (AIC). We address each of these sub-tasks with separate components without backward feedback between sub-tasks. However, the use of multiple parsers at the beginning of the process, and re-ranking at the end, contribute beneficial stochastic aspects to the system. Figure 1 summa-rizes the system architecture. We detail the parsing ∗All authors contributed equally to this work