- home
- Advanced Search
- Digital Humanities and Cultural Heritage
- 2017-2021
- Publications
- Research data
- Other research products
- Contribution for newspaper or weekl...
- GB
- OpenAIRE
- Digital Humanities and Cultural Heritage
- 2017-2021
- Publications
- Research data
- Other research products
- Contribution for newspaper or weekl...
- GB
- OpenAIRE
Loading
description Publicationkeyboard_double_arrow_right Conference object 2020 United KingdomPublisher:International Committee on Computational Linguistics Funded by:EC | M and MEC| M and MAuthors: Mitchell, Jeffrey J; Bowers, Jeffrey S;Mitchell, Jeffrey J; Bowers, Jeffrey S;Recently, domain-general recurrent neural networks, without explicit linguistic inductive biases, have been shown to successfully reproduce a range of human language behaviours, such as accurately predicting number agreement between nouns and verbs. We show that such networks will also learn number agreement within unnatural sentence structures, i.e. structures that are not found within any natural languages and which humans struggle to process. These results suggest that the models are learning from their input in a manner that is substantially different from human language acquisition, and we undertake an analysis of how the learned knowledge is stored in the weights of the network. We find that while the model has an effective understanding of singular versus plural for individual sentences, there is a lack of a unified concept of number agreement connecting these processes across the full range of inputs. Moreover, the weights handling natural and unnatural structures overlap substantially, in a way that underlines the non-human-like nature of the knowledge learned by the network.
Explore Bristol Rese... arrow_drop_down Explore Bristol Research; OpenAIREContribution for newspaper or weekly magazine . Conference object . 2020https://doi.org/10.18653/v1/20...Other literature type . Conference object . 2020 . Peer-reviewedadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/2020.coling-main.451&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.euAccess RoutesGreen hybrid 5 citations 5 popularity Top 10% influence Average impulse Average Powered by BIP!
more_vert Explore Bristol Rese... arrow_drop_down Explore Bristol Research; OpenAIREContribution for newspaper or weekly magazine . Conference object . 2020https://doi.org/10.18653/v1/20...Other literature type . Conference object . 2020 . Peer-reviewedadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/2020.coling-main.451&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.eudescription Publicationkeyboard_double_arrow_right Conference object , Article , Contribution for newspaper or weekly magazine , Other literature type 2018 France, Switzerland, United KingdomPublisher:Association for Computational Linguistics (ACL) Funded by:EC | SUMMA, EC | HimL, EC | TraMOOC +2 projectsEC| SUMMA ,EC| HimL ,EC| TraMOOC ,SNSF| Rich Context in Neural Machine Translation ,SNSF| Dating structural fabric development using high spatial resolution 40Ar/39Ar geochronology: a Combined Filed and Experimental ApproachAuthors: Bawden, Rachel; Sennrich, Rico; Birch, Alexandra; Haddow, Barry;Bawden, Rachel; Sennrich, Rico; Birch, Alexandra; Haddow, Barry;For machine translation to tackle discourse phenomena, models must have access to extra-sentential linguistic context. There has been recent interest in modelling context in neural machine translation (NMT), but models have been principally evaluated with standard automatic metrics, poorly adapted to evaluating discourse phenomena. In this article, we present hand-crafted, discourse test sets, designed to test the models' ability to exploit previous source and target sentences. We investigate the performance of recently proposed multi-encoder NMT models trained on subtitles for English to French. We also explore a novel way of exploiting context from the previous sentence. Despite gains using BLEU, multi-encoder models give limited improvement in the handling of discourse phenomena: 50% accuracy on our coreference test set and 53.5% for coherence/cohesion (compared to a non-contextual baseline of 50%). A simple strategy of decoding the concatenation of the previous and current sentence leads to good performance, and our novel strategy of multi-encoding and decoding of two sentences leads to the best performance (72.5% for coreference and 57% for coherence/cohesion), highlighting the importance of target-side context. Comment: Final version of paper to appear in Proceedings of NAACL 2018
OpenAIRE; Hal-Didero... arrow_drop_down OpenAIRE; Hal-DiderotConference object . 2018Hyper Article en Ligne; Mémoires en Sciences de l'Information et de la CommunicationOther literature type . Conference object . 2018Full-Text: https://hal.science/hal-01800739/documentZurich Open Repository and ArchiveConference object . 2018License: CC BYData sources: Zurich Open Repository and ArchiveEdinburgh Research ExplorerContribution for newspaper or weekly magazine . 2018Data sources: Edinburgh Research ExplorerarXiv.org e-Print ArchiveOther literature type . Preprint . 2017Data sources: arXiv.org e-Print Archivehttps://doi.org/10.48550/arxiv...Article . 2017License: arXiv Non-Exclusive DistributionData sources: Dataciteadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/n18-1118&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.euAccess RoutesGreen hybrid 89 citations 89 popularity Top 1% influence Top 10% impulse Top 1% Powered by BIP!
more_vert OpenAIRE; Hal-Didero... arrow_drop_down OpenAIRE; Hal-DiderotConference object . 2018Hyper Article en Ligne; Mémoires en Sciences de l'Information et de la CommunicationOther literature type . Conference object . 2018Full-Text: https://hal.science/hal-01800739/documentZurich Open Repository and ArchiveConference object . 2018License: CC BYData sources: Zurich Open Repository and ArchiveEdinburgh Research ExplorerContribution for newspaper or weekly magazine . 2018Data sources: Edinburgh Research ExplorerarXiv.org e-Print ArchiveOther literature type . Preprint . 2017Data sources: arXiv.org e-Print Archivehttps://doi.org/10.48550/arxiv...Article . 2017License: arXiv Non-Exclusive DistributionData sources: Dataciteadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/n18-1118&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.eudescription Publicationkeyboard_double_arrow_right Conference object , Preprint , Article , Contribution for newspaper or weekly magazine 2017 United KingdomPublisher:Association for Computational Linguistics (ACL) Funded by:EC | SUMMA, EC | QT21EC| SUMMA ,EC| QT21Authors: Sennrich, Rico;Sennrich, Rico;Analysing translation quality in regards to specific linguistic phenomena has historically been difficult and time-consuming. Neural machine translation has the attractive property that it can produce scores for arbitrary translations, and we propose a novel method to assess how well NMT systems model specific linguistic phenomena such as agreement over long distances, the production of novel words, and the faithful translation of polarity. The core idea is that we measure whether a reference translation is more probable under a NMT model than a contrastive translation which introduces a specific type of error. We present LingEval97, a large-scale data set of 97000 contrastive translation pairs based on the WMT English->German translation task, with errors automatically created with simple rules. We report results for a number of systems, and find that recently introduced character-level NMT systems perform better at transliteration than models with byte-pair encoding (BPE) segmentation, but perform more poorly at morphosyntactic agreement, and translating discontiguous units of meaning. accepted at EACL 2017 (v3: minor fix to table 6 description)
https://www.aclweb.o... arrow_drop_down Edinburgh Research ExplorerContribution for newspaper or weekly magazine . 2017Data sources: Edinburgh Research ExplorerarXiv.org e-Print ArchiveOther literature type . Preprint . 2016Data sources: arXiv.org e-Print Archivehttps://doi.org/10.48550/arxiv...Article . 2016License: arXiv Non-Exclusive DistributionData sources: Dataciteadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/e17-2060&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.euAccess RoutesGreen hybrid 12 citations 12 popularity Top 10% influence Average impulse Average Powered by BIP!
more_vert https://www.aclweb.o... arrow_drop_down Edinburgh Research ExplorerContribution for newspaper or weekly magazine . 2017Data sources: Edinburgh Research ExplorerarXiv.org e-Print ArchiveOther literature type . Preprint . 2016Data sources: arXiv.org e-Print Archivehttps://doi.org/10.48550/arxiv...Article . 2016License: arXiv Non-Exclusive DistributionData sources: Dataciteadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/e17-2060&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.eu
Loading
description Publicationkeyboard_double_arrow_right Conference object 2020 United KingdomPublisher:International Committee on Computational Linguistics Funded by:EC | M and MEC| M and MAuthors: Mitchell, Jeffrey J; Bowers, Jeffrey S;Mitchell, Jeffrey J; Bowers, Jeffrey S;Recently, domain-general recurrent neural networks, without explicit linguistic inductive biases, have been shown to successfully reproduce a range of human language behaviours, such as accurately predicting number agreement between nouns and verbs. We show that such networks will also learn number agreement within unnatural sentence structures, i.e. structures that are not found within any natural languages and which humans struggle to process. These results suggest that the models are learning from their input in a manner that is substantially different from human language acquisition, and we undertake an analysis of how the learned knowledge is stored in the weights of the network. We find that while the model has an effective understanding of singular versus plural for individual sentences, there is a lack of a unified concept of number agreement connecting these processes across the full range of inputs. Moreover, the weights handling natural and unnatural structures overlap substantially, in a way that underlines the non-human-like nature of the knowledge learned by the network.
Explore Bristol Rese... arrow_drop_down Explore Bristol Research; OpenAIREContribution for newspaper or weekly magazine . Conference object . 2020https://doi.org/10.18653/v1/20...Other literature type . Conference object . 2020 . Peer-reviewedadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/2020.coling-main.451&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.euAccess RoutesGreen hybrid 5 citations 5 popularity Top 10% influence Average impulse Average Powered by BIP!
more_vert Explore Bristol Rese... arrow_drop_down Explore Bristol Research; OpenAIREContribution for newspaper or weekly magazine . Conference object . 2020https://doi.org/10.18653/v1/20...Other literature type . Conference object . 2020 . Peer-reviewedadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/2020.coling-main.451&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.eudescription Publicationkeyboard_double_arrow_right Conference object , Article , Contribution for newspaper or weekly magazine , Other literature type 2018 France, Switzerland, United KingdomPublisher:Association for Computational Linguistics (ACL) Funded by:EC | SUMMA, EC | HimL, EC | TraMOOC +2 projectsEC| SUMMA ,EC| HimL ,EC| TraMOOC ,SNSF| Rich Context in Neural Machine Translation ,SNSF| Dating structural fabric development using high spatial resolution 40Ar/39Ar geochronology: a Combined Filed and Experimental ApproachAuthors: Bawden, Rachel; Sennrich, Rico; Birch, Alexandra; Haddow, Barry;Bawden, Rachel; Sennrich, Rico; Birch, Alexandra; Haddow, Barry;For machine translation to tackle discourse phenomena, models must have access to extra-sentential linguistic context. There has been recent interest in modelling context in neural machine translation (NMT), but models have been principally evaluated with standard automatic metrics, poorly adapted to evaluating discourse phenomena. In this article, we present hand-crafted, discourse test sets, designed to test the models' ability to exploit previous source and target sentences. We investigate the performance of recently proposed multi-encoder NMT models trained on subtitles for English to French. We also explore a novel way of exploiting context from the previous sentence. Despite gains using BLEU, multi-encoder models give limited improvement in the handling of discourse phenomena: 50% accuracy on our coreference test set and 53.5% for coherence/cohesion (compared to a non-contextual baseline of 50%). A simple strategy of decoding the concatenation of the previous and current sentence leads to good performance, and our novel strategy of multi-encoding and decoding of two sentences leads to the best performance (72.5% for coreference and 57% for coherence/cohesion), highlighting the importance of target-side context. Comment: Final version of paper to appear in Proceedings of NAACL 2018
OpenAIRE; Hal-Didero... arrow_drop_down OpenAIRE; Hal-DiderotConference object . 2018Hyper Article en Ligne; Mémoires en Sciences de l'Information et de la CommunicationOther literature type . Conference object . 2018Full-Text: https://hal.science/hal-01800739/documentZurich Open Repository and ArchiveConference object . 2018License: CC BYData sources: Zurich Open Repository and ArchiveEdinburgh Research ExplorerContribution for newspaper or weekly magazine . 2018Data sources: Edinburgh Research ExplorerarXiv.org e-Print ArchiveOther literature type . Preprint . 2017Data sources: arXiv.org e-Print Archivehttps://doi.org/10.48550/arxiv...Article . 2017License: arXiv Non-Exclusive DistributionData sources: Dataciteadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/n18-1118&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.euAccess RoutesGreen hybrid 89 citations 89 popularity Top 1% influence Top 10% impulse Top 1% Powered by BIP!
more_vert OpenAIRE; Hal-Didero... arrow_drop_down OpenAIRE; Hal-DiderotConference object . 2018Hyper Article en Ligne; Mémoires en Sciences de l'Information et de la CommunicationOther literature type . Conference object . 2018Full-Text: https://hal.science/hal-01800739/documentZurich Open Repository and ArchiveConference object . 2018License: CC BYData sources: Zurich Open Repository and ArchiveEdinburgh Research ExplorerContribution for newspaper or weekly magazine . 2018Data sources: Edinburgh Research ExplorerarXiv.org e-Print ArchiveOther literature type . Preprint . 2017Data sources: arXiv.org e-Print Archivehttps://doi.org/10.48550/arxiv...Article . 2017License: arXiv Non-Exclusive DistributionData sources: Dataciteadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/n18-1118&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.eudescription Publicationkeyboard_double_arrow_right Conference object , Preprint , Article , Contribution for newspaper or weekly magazine 2017 United KingdomPublisher:Association for Computational Linguistics (ACL) Funded by:EC | SUMMA, EC | QT21EC| SUMMA ,EC| QT21Authors: Sennrich, Rico;Sennrich, Rico;Analysing translation quality in regards to specific linguistic phenomena has historically been difficult and time-consuming. Neural machine translation has the attractive property that it can produce scores for arbitrary translations, and we propose a novel method to assess how well NMT systems model specific linguistic phenomena such as agreement over long distances, the production of novel words, and the faithful translation of polarity. The core idea is that we measure whether a reference translation is more probable under a NMT model than a contrastive translation which introduces a specific type of error. We present LingEval97, a large-scale data set of 97000 contrastive translation pairs based on the WMT English->German translation task, with errors automatically created with simple rules. We report results for a number of systems, and find that recently introduced character-level NMT systems perform better at transliteration than models with byte-pair encoding (BPE) segmentation, but perform more poorly at morphosyntactic agreement, and translating discontiguous units of meaning. accepted at EACL 2017 (v3: minor fix to table 6 description)
https://www.aclweb.o... arrow_drop_down Edinburgh Research ExplorerContribution for newspaper or weekly magazine . 2017Data sources: Edinburgh Research ExplorerarXiv.org e-Print ArchiveOther literature type . Preprint . 2016Data sources: arXiv.org e-Print Archivehttps://doi.org/10.48550/arxiv...Article . 2016License: arXiv Non-Exclusive DistributionData sources: Dataciteadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/e17-2060&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.euAccess RoutesGreen hybrid 12 citations 12 popularity Top 10% influence Average impulse Average Powered by BIP!
more_vert https://www.aclweb.o... arrow_drop_down Edinburgh Research ExplorerContribution for newspaper or weekly magazine . 2017Data sources: Edinburgh Research ExplorerarXiv.org e-Print ArchiveOther literature type . Preprint . 2016Data sources: arXiv.org e-Print Archivehttps://doi.org/10.48550/arxiv...Article . 2016License: arXiv Non-Exclusive DistributionData sources: Dataciteadd ClaimPlease grant OpenAIRE to access and update your ORCID works.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.This Research product is the result of merged Research products in OpenAIRE.
You have already added works in your ORCID record related to the merged Research product.All Research productsarrow_drop_down <script type="text/javascript"> <!-- document.write('<div id="oa_widget"></div>'); document.write('<script type="text/javascript" src="https://www.openaire.eu/index.php?option=com_openaire&view=widget&format=raw&projectId=10.18653/v1/e17-2060&type=result"></script>'); --> </script>
For further information contact us at helpdesk@openaire.eu