Efficient decoding for syntactic parsing has become a necessary research area as statistical grammars grow in accuracy and size and as more NLP applications leverage syntactic analyses. We review prior methods for pruning and then present a new framework that unifies their strengths into a single approach. Using a log linear model, we learn the optimal beam-search pruning parameters for each CYK chart cell, effectively predicting the most promising areas of the model space to explore. We demonstrate that our method is faster than coarse-to-fine pruning, exemplified in both the Charniak and Berkeley parsers, by empirically comparing our parser to the Berkeley parser using the same grammar and under identical operating conditions. 1

Semantic role labeling (SRL) and word sense disambiguation (WSD) are two fundamental tasks in natural language processing to find a sentence-level semantic representation. To date, they have mostly been modeled in isolation. However, this approach neglects logical constraints between them. We therefore exploit some pipeline systems which verify the automatic all word sense disambiguation could help the semantic role labeling and vice versa. We further propose a Markov logic model that jointly labels semantic roles and disambiguates all word senses. By evaluating our model on the OntoNotes 3.0 data, we show that this joint approach leads to a higher performance for word sense disambiguation and semantic role labeling than those pipeline approaches. 1

We describe a semantic role labeling system that makes primary use of CCG-based features. Most previously developed systems are CFG-based and make extensive use of a treepath feature, which suffers from data sparsity due to its use of explicit tree configurations. CCG affords ways to augment treepathbased features to overcome these data sparsity issues. By adding features over CCG wordword dependencies and lexicalized verbal subcategorization frames (“supertags”), we can obtain an F-score that is substantially better than a previous CCG-based SRL system and competitive with the current state of the art. A manual error analysis reveals that parser errors account for many of the errors of our system. This analysis also suggests that simultaneous incremental parsing and semantic role labeling may lead to performance gains in both tasks. 1

reproduced, without authorization, under the conditions for Fair Dealing. Therefore, limited reproduction of this work for the purposes of private study, research, criticism, review and news reporting is likely to be in accordance with the law, particularly if cited appropriately. APPROVAL Name: Degree:

In many NLP systems, there is a unidirectional flow of information in which a parser supplies input to a semantic role labeler. In this paper, we build a system that allows information to flow in both directions. We make use of semantic role predictions in choosing a single-best parse. This process relies on an averaged perceptron model to distinguish likely semantic roles from erroneous ones. Our system penalizes parses that give rise to low-scoring semantic roles. To explore the consequences of this we perform two experiments. First, we use a baseline generative model to produce n-best parses, which are then re-ordered by our semantic model. Second, we use a modified version of our semantic role labeler to predict semantic roles at parse time. The performance of this modified labeler is weaker than that of our best full SRL, because it is restricted to features that can be computed directly from the parser’s packed chart. For both experiments, the resulting semantic predictions are then used to select parses. Finally, we feed the selected parses produced by each experiment to the full version of our semantic role labeler. We find that SRL performance can be improved over this baseline by selecting parses with likely semantic roles. 1

Natural language systems trained on labeled data from one domain do not perform well on other domains. Most adaptation algorithms proposed in the literature train a new model for the new domain using unlabeled data. However, it is time consuming to retrain big models or pipeline systems. Moreover, the domain of a new target sentence may not be known, and one may not have significant amount of unlabeled data for every new domain. To pursue the goal of an Open Domain NLP (train once, test anywhere), we propose ADUT (ADaptation Using label-preserving Transformation), an approach that avoids the need for retraining and does not require knowledge of the new domain, or any data from it. Our approach applies simple label-preserving transformations to the target text so that the transformed text is more similar to the training domain; it then applies the existing model on the transformed sentences and combines the predictions to produce the desired prediction on the target text. We instantiate ADUT for the case of Semantic Role Labeling (SRL) and show that it compares favorably with approaches that retrain their model on the target domain. Specifically, this “on the fly ” adaptation approach yields 13 % error reduction for a single parse system when adapting from the news wire text to fiction. 1