The PASCAL Recognising Textual Entailment Challenge

Dagan, Ido; Glickman, Oren; Magnini, Bernardo

doi:10.1007/11736790_9

Cited by 1,117 publications

(905 citation statements)

References 7 publications

Supporting

Mentioning

815

Contrasting

Unclassified

Order By: Relevance

“…By way of example, research in Textual Entailment is supported by the availability of several annotated datasets. These resources typically consist of sets of T-H pairs manually annotated with a Boolean value to indicate whether or not H is entailed by T. In the current paper, datasets RTE1 [4], RTE-2 [5], and RTE-3 [6] are used to evaluate the impact of coreference resolution on automatic RTE. We have opted to use such publicly available resources in spite of the fact that, as a result, we had to resort to the exploitation of different data for every evaluation/application in contrast to our previous experiments where we benefited from a common corpus.…”

Section: Introductionmentioning

confidence: 99%

Coreference Resolution: To What Extent Does It Help NLP Applications?

Mitkov

Evans

Orăsan

et al. 2012

Text, Speech and Dialogue

View full text Add to dashboard Cite

This paper describes a study of the impact of coreference resolution on NLP applications. Further to our previous study [1], in which we investigated whether anaphora resolution could be beneficial to NLP applications, we now seek to establish whether a different, but related task -that of coreference resolution, could improve the performance of three NLP applications: text summarisation, recognising textual entailment and text classification. The study discusses experiments in which the aforementioned applications were implemented in two versions, one in which the BART coreference resolution system was integrated and one in which it was not, and then tested in processing input text. The paper discusses the results obtained.

show abstract

Section: Introductionmentioning

confidence: 99%

Coreference Resolution: To What Extent Does It Help NLP Applications?

Mitkov

Evans

Orăsan

et al. 2012

Text, Speech and Dialogue

View full text Add to dashboard Cite

show abstract

“…a human annotated corpus), in which implicit information does not appear. In textual entailment (Dagan et al, 2005) the measurement problem is similar, however they address this in evaluations by always making the determination based on pairs of text passages. So we can show improvement in recall by selecting meaningful queries and determining if and how reasoning improves the recall for each query, but measuring recall improvements in the KB itself is more difficult.…”

Section: Techniques For Improving Recallmentioning

confidence: 99%

Towards Knowledge Acquisition from Information Extraction

Welty

Murdock

2006

Lecture Notes in Computer Science

View full text Add to dashboard Cite

Abstract. In our research to use information extraction to help populate the semantic web, we have encountered significant obstacles to interoperability between the technologies. We believe these obstacles to be endemic to the basic paradigms, and not quirks of the specific implementations we have worked with. In particular, we identify five dimensions of interoperability that must be addressed to successfully populate semantic web knowledge bases from information extraction systems that are suitable for reasoning. We call the task of transforming IE data into knowledge-bases knowledge integration, and briefly present a framework called KITE in which we are exploring these dimensions. Finally, we report on the initial results of an experiment in which the knowledge integration process uses the deeper semantics of OWL ontologies to improve the precision of relation extraction from text.

show abstract

“…While the most prominent forum using textual entailment is the Recognizing Textual Entailment (RTE) challenge (Dagan et al, 2005), the RTE datasets do not test the phenomena in which we are interested. For example, in order to evaluate our system's ability to determine word meaning in context, the RTE pair would have to specifically test word sense confusion by having a word's context in the hypothesis be different from the context of the premise.…”

Section: Introductionmentioning

confidence: 99%

A Formal Approach to Linking Logical Form and Vector-Space Lexical Semantics

Garrette

Erk

Mooney

2014

Text, Speech and Language Technology

View full text Add to dashboard Cite

First-order logic provides a powerful and flexible mechanism for representing natural language semantics. However, it is an open question of how best to integrate it with uncertain, weighted knowledge, for example regarding word meaning. This paper describes a mapping between predicates of logical form and points in a vector space. This mapping is then used to project distributional inferences to inference rules in logical form. We then describe first steps of an approach that uses this mapping to recast first-order semantics into the probabilistic models that are part of Statistical Relational AI. Specifically, we show how Discourse Representation Structures can be combined with distributional models for word meaning inside a Markov Logic Network and used to successfully perform inferences that take advantage of logical concepts such as negation and factivity as well as weighted information on word meaning in context.

show abstract

The PASCAL Recognising Textual Entailment Challenge

Cited by 1,117 publications

References 7 publications

Coreference Resolution: To What Extent Does It Help NLP Applications?

Coreference Resolution: To What Extent Does It Help NLP Applications?

Towards Knowledge Acquisition from Information Extraction

A Formal Approach to Linking Logical Form and Vector-Space Lexical Semantics

Contact Info

Product

Resources

About