Please use this identifier to cite or link to this item:
Full metadata record
DC FieldValueLanguage
dc.identifier.citationTanti, M., Gatt, A., & Camilleri, K. P. (2017). What is the role of recurrent neural networks (RNNs) in an image caption generator?. arXiv preprint arXiv:1708.02043.en_GB
dc.description.abstractIn neural image captioning systems, a recurrent neural network (RNN) is typically viewed as the primary `generation' component. This view suggests that the image features should be `injected' into the RNN. This is in fact the dominant view in the literature. Alternatively, the RNN can instead be viewed as only encoding the previously generated words. This view suggests that the RNN should only be used to encode linguistic features and that only the final representation should be `merged' with the image features at a later stage. This paper compares these two architectures. We find that, in general, late merging outperforms injection, suggesting that RNNs are better viewed as encoders, rather than generators.en_GB
dc.publisherCornell Universityen_GB
dc.subjectComputational linguisticsen_GB
dc.subjectImage analysisen_GB
dc.subjectNatural language processing (Computer science)en_GB
dc.subjectLinguistic analysis (Linguistics)en_GB
dc.subjectCorpora (Linguistics)en_GB
dc.titleWhat is the role of recurrent neural networks (RNNs) in an image caption generator?en_GB
dc.rights.holderThe copyright of this work belongs to the author(s)/publisher. The rights of this work are as defined by the appropriate Copyright Legislation or as modified by any successive legislation. Users may access this work and can make use of the information contained in accordance with the Copyright Legislation provided that the author must be properly acknowledged. Further distribution or reproduction in any format is prohibited without the prior permission of the copyright holderen_GB
dc.contributor.creatorTanti, Marc-
dc.contributor.creatorGatt, Albert-
dc.contributor.creatorCamilleri, Kenneth P.-
Appears in Collections:Scholarly Works - FacEngSCE
Scholarly Works - InsLin

Files in This Item:
File Description SizeFormat 
inlg2017-rnn.pdf163.47 kBAdobe PDFView/Open

Items in OAR@UM are protected by copyright, with all rights reserved, unless otherwise indicated.