Show simple item record

dc.contributor.authorSuhr, Alane
dc.contributor.authorIyer, Srinivasan
dc.contributor.authorArtzi, Yoav
dc.date.accessioned2019-10-28T20:21:19Z
dc.date.available2019-10-28T20:21:19Z
dc.date.created2018-09-05 00:44
dc.date.issued2018-04-18
dc.identifieroai:arXiv.org:1804.06868
dc.identifierhttp://arxiv.org/abs/1804.06868
dc.identifier.urihttp://hdl.handle.net/20.500.12424/2486392
dc.description.abstractWe propose a context-dependent model to map utterances within an interaction to executable formal queries. To incorporate interaction history, the model maintains an interaction-level encoder that updates after each turn, and can copy sub-sequences of previously predicted queries during generation. Our approach combines implicit and explicit modeling of references between utterances. We evaluate our model on the ATIS flight planning interactions, and demonstrate the benefits of modeling context and explicit references.
dc.description.abstractComment: NAACL-HLT 2018
dc.subjectComputer Science - Computation and Language
dc.titleLearning to Map Context-Dependent Sentences to Executable Formal Queries
dc.typetext
ge.collectioncodeOAIDATA
ge.dataimportlabelOAI metadata object
ge.identifier.legacyglobethics:15135108
ge.identifier.permalinkhttps://www.globethics.net/gel/15135108
ge.lastmodificationdate2018-09-05 00:44
ge.lastmodificationuseradmin@pointsoftware.ch (import)
ge.submissions0
ge.oai.exportid149801
ge.oai.repositoryid58
ge.oai.setnameComputer Science
ge.oai.setspeccs
ge.oai.streamid2
ge.setnameGlobeEthicsLib
ge.setspecglobeethicslib
ge.linkhttp://arxiv.org/abs/1804.06868


This item appears in the following Collection(s)

Show simple item record