Matches in SemOpenAlex for { <https://semopenalex.org/work/W2891879983> ?p ?o ?g. }
Showing items 1 to 63 of
63
with 100 items per page.
- W2891879983 abstract "The ease of constructing effective neural networks has resulted in a large number of varying architectures iteratively improving performance on a task. Due to the nature of these models being black boxes, standard weight inspection is difficult. We propose a probe based methodology to evaluate what information is important or extraneous at each level of a network. We input natural language processing datasets into a trained answer passage neural network. Each layer of the neural network is used as input into a unique classifier, or probe, to correctly label that input with respect to a natural language processing task, probing the internal representations for information. Using this approach, we analyze the information relevant to retrieving answer passages from the perspective of information needed for part of speech tagging, named entity retrieval, sentiment classification, and textual entailment. We show a significant information need difference between two seemingly similar question answering collections, and demonstrate that passage retrieval and textual entailment share a common information space, while POS and NER information is used only at a compositional level in the lower layers of an information retrieval model. Lastly, we demonstrate that incorporating this information into a multitask environment is correlated to the information retained by these models during the probe inspection phase." @default.
- W2891879983 created "2018-09-27" @default.
- W2891879983 creator A5023365550 @default.
- W2891879983 creator A5047094005 @default.
- W2891879983 creator A5082506991 @default.
- W2891879983 date "2018-09-10" @default.
- W2891879983 modified "2023-10-16" @default.
- W2891879983 title "Understanding the Representational Power of Neural Retrieval Models Using NLP Tasks" @default.
- W2891879983 cites W1010415138 @default.
- W2891879983 cites W1966443646 @default.
- W2891879983 cites W2093647425 @default.
- W2891879983 cites W2144578941 @default.
- W2891879983 cites W2296283641 @default.
- W2891879983 cites W2341132943 @default.
- W2891879983 cites W2515565210 @default.
- W2891879983 cites W3122775348 @default.
- W2891879983 doi "https://doi.org/10.1145/3234944.3234959" @default.
- W2891879983 hasPublicationYear "2018" @default.
- W2891879983 type Work @default.
- W2891879983 sameAs 2891879983 @default.
- W2891879983 citedByCount "3" @default.
- W2891879983 countsByYear W28918799832019 @default.
- W2891879983 countsByYear W28918799832020 @default.
- W2891879983 crossrefType "proceedings-article" @default.
- W2891879983 hasAuthorship W2891879983A5023365550 @default.
- W2891879983 hasAuthorship W2891879983A5047094005 @default.
- W2891879983 hasAuthorship W2891879983A5082506991 @default.
- W2891879983 hasBestOaLocation W28918799831 @default.
- W2891879983 hasConcept C121332964 @default.
- W2891879983 hasConcept C154945302 @default.
- W2891879983 hasConcept C163258240 @default.
- W2891879983 hasConcept C195818886 @default.
- W2891879983 hasConcept C199360897 @default.
- W2891879983 hasConcept C204321447 @default.
- W2891879983 hasConcept C41008148 @default.
- W2891879983 hasConcept C62520636 @default.
- W2891879983 hasConceptScore W2891879983C121332964 @default.
- W2891879983 hasConceptScore W2891879983C154945302 @default.
- W2891879983 hasConceptScore W2891879983C163258240 @default.
- W2891879983 hasConceptScore W2891879983C195818886 @default.
- W2891879983 hasConceptScore W2891879983C199360897 @default.
- W2891879983 hasConceptScore W2891879983C204321447 @default.
- W2891879983 hasConceptScore W2891879983C41008148 @default.
- W2891879983 hasConceptScore W2891879983C62520636 @default.
- W2891879983 hasFunder F4320332236 @default.
- W2891879983 hasFunder F4320338294 @default.
- W2891879983 hasLocation W28918799831 @default.
- W2891879983 hasOpenAccess W2891879983 @default.
- W2891879983 hasPrimaryLocation W28918799831 @default.
- W2891879983 hasRelatedWork W1515542156 @default.
- W2891879983 hasRelatedWork W1552159754 @default.
- W2891879983 hasRelatedWork W2131420137 @default.
- W2891879983 hasRelatedWork W2148757832 @default.
- W2891879983 hasRelatedWork W2293457016 @default.
- W2891879983 hasRelatedWork W2368651715 @default.
- W2891879983 hasRelatedWork W2611614995 @default.
- W2891879983 hasRelatedWork W2789919619 @default.
- W2891879983 hasRelatedWork W3107474891 @default.
- W2891879983 hasRelatedWork W3169305685 @default.
- W2891879983 isParatext "false" @default.
- W2891879983 isRetracted "false" @default.
- W2891879983 magId "2891879983" @default.
- W2891879983 workType "article" @default.