Methods for using textual entailment in open-domain question answering

Harabagiu, Sanda M.; Hickl, Andrew

doi:10.3115/1220175.1220289

Cited by 142 publications

(88 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…A fundamental motivation and long standing goal of the paraphrasing and RTE communities has been to cast various semantic applications as paraphrasing/textual entailment (Dagan et al, 2013). While it has been shown that paraphrasing methods are useful for question answering (Harabagiu and Hickl, 2006) and relation extraction (Romano et al, 2006), this is, to the best of our knowledge, the first paper to perform semantic parsing through paraphrasing. Our paraphrase model emphasizes simplicity and efficiency, but the framework is agnostic to the internals of the paraphrase method.…”

Section: Discussionmentioning

confidence: 99%

Semantic Parsing via Paraphrasing

Berant¹,

Liang²

2014

Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

473

416

View full text Add to dashboard Cite

A central challenge in semantic parsing is handling the myriad ways in which knowledge base predicates can be expressed. Traditionally, semantic parsers are trained primarily from text paired with knowledge base information. Our goal is to exploit the much larger amounts of raw text not tied to any knowledge base. In this paper, we turn semantic parsing on its head. Given an input utterance, we first use a simple method to deterministically generate a set of candidate logical forms with a canonical realization in natural language for each. Then, we use a paraphrase model to choose the realization that best paraphrases the input, and output the corresponding logical form. We present two simple paraphrase models, an association model and a vector space model, and train them jointly from question-answer pairs. Our system PARASEMPRE improves stateof-the-art accuracies on two recently released question-answering datasets.

show abstract

Section: Discussionmentioning

confidence: 99%

Semantic Parsing via Paraphrasing

Berant¹,

Liang²

2014

Proceedings of the 52nd Annual Meeting of the Association for Computational Linguistics (Volume 1: Long Papers)

473

416

View full text Add to dashboard Cite

show abstract

“…Punyakanok et al (2004) compute the tree edit distance between the dependency trees of the question and answer, and Bouma et al (2005) use deep syntactic parsing and distributional similarities from external corpora). Even though results to the RTE task in general were modest with accuracy scores between 50-60%, for specific task settings, they could bring accuracy gains: Harabagiu (2006) report an increase in performance from 30.6% to 42.7% on an open-domain QA task.…”

Section: Related Workmentioning

confidence: 99%

Mining Social Science Publications for Survey Variables

Zielinski

Mutschke²

2017

Proceedings of the Second Workshop on NLP and Computational Social Science

View full text Add to dashboard Cite

Research in Social Science is usually based on survey data where individual research questions relate to observable concepts (variables). However, due to a lack of standards for data citations a reliable identification of the variables used is often difficult. In this paper, we present a work-in-progress study that seeks to provide a solution to the variable detection task based on supervised machine learning algorithms, using a linguistic analysis pipeline to extract a rich feature set, including terminological concepts and similarity metric scores. Further, we present preliminary results on a small dataset that has been specifically designed for this task, yielding a significant increase in performance over the random baseline.

show abstract

“…Some recent works [5] have shown how the use of textual entailment can improve the accuracy of QA systems. Our aim was to obtain evidences of this improvement in a comparative and shared evaluation.…”

Section: Evaluation Of the Answer Validation Exercisementioning

confidence: 99%

“…In some sense, systems must emulate human assessment of QA responses and decide whether an answer is correct or not according to a given text. This automatic Answer Validation is expected to be useful for improving QA systems performance [5]. However, the evaluation methodology in AVE 2006 did not permit to quantify this improvement and thus, the exercise has been modified in AVE 2007.…”

Section: Introductionmentioning

confidence: 99%

Overview of the Answer Validation Exercise 2008

Rodrigo

Peñas

Verdejo

2009

Lecture Notes in Computer Science

View full text Add to dashboard Cite

The Answer Validation Exercise at the Cross Language Evaluation Forum is aimed at developing systems able to decide whether the answer of a Question Answering system is correct or not. We present here the exercise description, the changes in the evaluation methodology with respect to the first edition, and the results of this second edition (AVE 2007). The changes in the evaluation methodology had two objectives: the first one was to quantify the gain in performance when more sophisticated validation modules are introduced in QA systems. The second objective was to bring systems based on Textual Entailment to the Automatic Hypothesis Generation problem which is not part itself of the Recognising Textual Entailment (RTE) task but a need of the Answer Validation setting. 9 groups have participated with 16 runs in 4 different languages. Compared with the QA systems, the results show an evidence of the potential gain that more sophisticated AV modules introduce in the task of QA.

show abstract

Methods for using textual entailment in open-domain question answering

Cited by 142 publications

References 14 publications

Semantic Parsing via Paraphrasing

Semantic Parsing via Paraphrasing

Mining Social Science Publications for Survey Variables

Overview of the Answer Validation Exercise 2008

Contact Info

Product

Resources

About