Sentence compression holds promise for many applications ranging from summarization to subtitle generation. Our work views sentence compression as an optimization problem and uses integer linear programming (ILP) to infer globally optimal compressions in the presence of linguistically motivated constraints. We show how previous formulations of sentence compression can be recast as ILPs and extend these models with novel global constraints. Experimental results on written and spoken texts demonstrate improvements over state-of-the-art models. 1.

Although much work in NLP has focused on simply determining what a document means, we also must know whether or not to believe it. Fact-finding algorithms attempt to identify the “truth ” among competing claims in a corpus, but fail to take advantage of the user’s prior knowledge and presume that truth itself is universal and objective rather than subjective. We introduce a framework for incorporating prior knowledge into any factfinding algorithm, expressing both general “common-sense ” reasoning and specific facts already known to the user as first-order logic and translating this into a tractable linear program. As our results show, this approach scales well to even large problems, both reducing error and allowing the system to determine truth respective to the user rather than the majority. Additionally, we introduce three new fact-finding algorithms capable of outperforming existing fact-finders in many of our experiments. 1

There are major trends to advance the functionality of search engines to a more expressive semantic level. This is enabled by the advent of knowledge-sharing communities such as Wikipedia and the progress in automatically extracting entities and relationships from semistructured as well as natural-language Web sources. Recent endeavors of this kind include DBpedia, EntityCube, KnowItAll, ReadTheWeb, and our own YAGO-NAGA project (and others). The goal is to automatically construct and maintain a comprehensive knowledge base of facts about named entities, their semantic classes, and their mutual relations as well as temporal contexts, with high precision and high recall. This tutorial discusses state-ofthe-art methods, research opportunities, and open challenges along this avenue of knowledge harvesting.

Current approaches for semantic parsing take a supervised approach requiring a considerable amount of training data which is expensive and difficult to obtain. This supervision bottleneck is one of the major difficulties in scaling up semantic parsing. We argue that a semantic parser can be trained effectively without annotated data, and introduce an unsupervised learning algorithm. The algorithm takes a self training approach driven by confidence estimation. Evaluated over Geoquery, a standard dataset for this task, our system achieved 66 % accuracy, compared to 80 % of its fully supervised counterpart, demonstrating the promise of unsupervised approaches for this task. 1

We present a novel approach to relation extraction that integrates information across documents, performs global inference and requires no labelled text. In particular, we tackle relation extraction and entity identification jointly. We use distant supervision to train a factor graph model for relation extraction based on an existing knowledge base (Freebase, derived in parts from Wikipedia). For inference we run an efficient Gibbs sampler that leads to linear time joint inference. We evaluate our approach both for an indomain (Wikipedia) and a more realistic outof-domain (New York Times Corpus) setting. For the in-domain setting, our joint model leads to 4 % higher precision than an isolated local approach, but has no advantage over a pipeline. For the out-of-domain data, we benefit strongly from joint modelling, and observe improvements in precision of 13 % over the pipeline, and 15 % over the isolated baseline. 1

This paper studies textual inference by investigating comma structures, which are highly frequent elements whose major role in the extraction of semantic relations has not been hitherto recognized. We introduce the problem of comma resolution, defined as understanding the role of commas and extracting the relations they imply. We show the importance of the problem using examples from Textual Entailment tasks, and present A Sentence Transformation Rule Learner (ASTRL), a machine learning algorithm that uses a syntactic analysis of the sentence to learn sentence transformation rules that can then be used to extract relations. We have manually annotated a corpus identifying comma structures and relations they entail and experimented with both gold standard parses and parses created by a leading statistical parser, obtaining F-scores of 80.2 % and 70.4 % respectively. 1

We introduce several ideas that improve the performance of supervised information extraction systems with a pipeline architecture, when they are customized for new domains. We show that: (a) a combination of a sequence tagger with a rule-based approach for entity mention extraction yields better performance for both entity and relation mention extraction; (b) improving the identification of syntactic heads of entity mentions helps relation extraction; and (c) a deterministic inference engine captures some of the joint domain structure, even when introduced as a postprocessing step to a pipeline system. All in all, our contributions yield a 20 % relative increase in F1 score in a domain significantly different from the domains used during the development of our information extraction system. 1

Determining whether two terms in text have an ancestor relation (e.g. Toyota and car) or a sibling relation (e.g. Toyota and Honda) is an essential component of textual inference in NLP applications such as Question Answering, Summarization, and Recognizing Textual Entailment. Significant work has been done on developing stationary knowledge sources that could potentially support these tasks, but these resources often suffer from low coverage, noise, and are inflexible when needed to support terms that are not identical to those placed in them, making their use as general purpose background knowledge resources difficult. In this paper, rather than building a stationary hierarchical structure of terms and relations, we describe a system that, given two terms, determines the taxonomic relation between them using a machine learning-based approach that makes use of existing resources. Moreover, we develop a global constraint optimization inference process and use it to leverage an existing knowledge base also to enforce relational constraints among terms and thus improve the classifier predictions. Our experimental evaluation shows that our approach significantly outperforms other systems built upon existing well-known knowledge sources. 1

For many machine learning solutions to complex applications, there are significant performance advantages to decomposing the overall task into several simpler sequential stages, commonly referred to as a pipeline model. Typically, such scenarios are also characterized by high sample complexity, motivating the study of active learning for these situations. While most active learning research examines single predictions, we extend such work to applications which utilize pipelined predictions. Specifically, we present an adaptive strategy for combining local active learning strategies into one that minimizes the annotation requirements for the overall task. Empirical results for a three-stage entity and relation extraction system demonstrate a significant reduction in supervised data requirements when using the proposed method.

This paper introduces a novel constraint-driven learning framework for identifying named-entity (NE) transliterations. Traditional approaches to the problem of discovering transliterations depend heavily on correctly segmenting the target and the transliteration candidate and on and aligning these segments. In this work we propose to formulate the process of aligning segments as a constrained optimization problem. We consider the aligned segments as a latent feature representation and show how to infer an optimal latent representation and how to use it in order to learn an improved discriminative transliteration classifier. Our algorithm is an EM-like iterative algorithm that alternates between an optimization step for the latent representation and a learning step for the classifier’s parameters. We apply this method both in supervised and unsupervised settings, and show that our model can significantly outperform previous methods trained using considerably more resources. 1