Filters








145,858 Hits in 3.4 sec

Audio-Linguistic Embeddings for Spoken Sentences [article]

Albert Haque, Michelle Guo, Prateek Verma, Li Fei-Fei
2019 arXiv   pre-print
We propose spoken sentence embeddings which capture both acoustic and linguistic content.  ...  Ablation studies show that our embeddings can better model high-level acoustic concepts while retaining linguistic content.  ...  In this work, we investigate embeddings in the domain of spoken language processing and propose spoken sentence embeddings, capable of modeling both acoustic and linguistic content in a single latent code  ... 
arXiv:1902.07817v1 fatcat:nsgpzkgaqvcpxieuqvdytewhcq

Improving Disentangled Text Representation Learning with Information-Theoretic Guidance [article]

Pengyu Cheng, Martin Renqiang Min, Dinghan Shen, Christopher Malon, Yizhe Zhang, Yitong Li, Lawrence Carin
2022 arXiv   pre-print
By minimizing this upper bound, the proposed method induces style and content embeddings into two independent low-dimensional spaces.  ...  Experiments on both conditional text generation and text-style transfer demonstrate the high quality of our disentangled representation in terms of content and style preservation.  ...  The content embedding is designed to encapsulate the semantic meaning of a sentence.  ... 
arXiv:2006.00693v3 fatcat:l22eiy6ri5ghdjb5muq7rzvbru

Encouraging Paragraph Embeddings to Remember Sentence Identity Improves Classification

Tu Vu, Mohit Iyyer
2019 Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics  
In this paper, we investigate a state-of-the-art paragraph embedding method proposed by Zhang et al. (2017) and discover that it cannot reliably tell whether a given sentence occurs in the input paragraph  ...  This result motivates us to replace the reconstructionbased objective of Zhang et al. (2017) with our sentence content probe objective in a semisupervised setting.  ...  We thus design a probe task to measure the extent to which this sentence content property is captured in a paragraph embedding.  ... 
doi:10.18653/v1/p19-1638 dblp:conf/acl/VuI19 fatcat:hvshbxfpcjasreuik2gsbsc7fe

QLUT at SemEval-2017 Task 1: Semantic Textual Similarity Based on Word Embeddings

Fanqing Meng, Wenpeng Lu, Yuteng Zhang, Jinyong Cheng, Yuehan Du, Shuwang Han
2017 Proceedings of the 11th International Workshop on Semantic Evaluation (SemEval-2017)  
This paper reports the details of our submissions in the task 1 of SemEval 2017. This task aims at assessing the semantic textual similarity of two sentences or texts.  ...  in improving the performance of models.  ...  Acknowledgments The work described in this paper is mainly supported by Natural Science Foundation of China under Grant 61502259 and 61202244, Natural Science Foundation of Shandong Province under Grant  ... 
doi:10.18653/v1/s17-2020 dblp:conf/semeval/MengLZCDH17 fatcat:4detwfz2ujbxtcbfmoulbwz7ja

14 Syntactic structure, grammatical accuracy, and content in second-language writing: An analysis of skill learning and on-line processing [chapter]

Wendy S. Francis, Laura F. Romo, Rochel Gelman
2002 Advances in Psychology  
Written language samples were obtained from high school students enrolled in an intermediate-level English as a Second Language (ESL) class with embedded science instruction.  ...  This chapter examines second-language sentence production both as a skill and as an on-line process by measuring relationships among linguistic and content characteristics of language samples both between  ...  Content accuracy Embedded clause Prep.  ... 
doi:10.1016/s0166-4115(02)80017-6 fatcat:dwcjetlllnhxniybu7fywkcrbi

ON POSSIBILITIES OF MULTILINGUAL BERT MODEL FOR DETERMINING SEMANTIC SIMILARITIES OF THE NEWS CONTENT

S. Olizarenko, V. Argunov
2020 Системи управління навігації та зв'язку Збірник наукових праць  
An efficient method is proposed to define the semantic similarities of the multilingual news content in HIPSTO aggregated news feeds on the basis of the sentence embeddings using the first task of the  ...  In particular, the possibilities and ways of applying the multilingual BERT model to handle the problem of semantic likeness of news content have been investigated.  ...  A promising approach for finding semantic similarity in the analysis of multilingual news content is provided in the pre-trained multilingual model at the level of sentence embeddings.  ... 
doi:10.26906/sunz.2020.3.094 fatcat:mjrezk2ck5hjpiq3liba6wc5f4

Jointly Modeling Embedding and Translation to Bridge Video and Language [article]

Yingwei Pan, Tao Mei, Ting Yao, Houqiang Li, Yong Rui
2015 arXiv   pre-print
relationship between the semantics of the entire sentence and visual content.  ...  model for exploring the relationships between visual content and sentence semantics.  ...  and video content which is measured in the visualsemantic embedding.  ... 
arXiv:1505.01861v3 fatcat:tvpbppgjabaytl2xlsi2mrtxmi

Fine-grained Analysis of Sentence Embeddings Using Auxiliary Prediction Tasks [article]

Yossi Adi, Einat Kermany, Yonatan Belinkov, Ofer Lavi, Yoav Goldberg
2017 arXiv   pre-print
There is a lot of research interest in encoding variable length sentences into fixed length vectors, in a way that preserves the sentence meanings.  ...  We define prediction tasks around isolated aspects of sentence structure (namely sentence length, word content, and word order), and score representations by the ability to train a classifier to solve  ...  ., 2014; in reversing the input sentences and clipping gradients. Word vectors are initialized to random values.  ... 
arXiv:1608.04207v3 fatcat:xe3v7k3ggvc2nn3dev5kvtavq4

Analysis of Bag-of-n-grams Representation's Properties Based on Textual Reconstruction [article]

Qi Huang and Zhanghao Chen and Zijie Lu and Yuan Ye
2018 arXiv   pre-print
We then run prediction tasks (sen- tence length, word content, phrase content and word order) using the obtained representation to look into the specific type of information captured in the representation  ...  Our analysis demonstrates that bag-of-n-grams representa- tion does contain sentence structure level in- formation.  ...  Sentence reconstruction gives us general-purposed embeddings, and we then test the obtained embeddings on prediction tasks including length, word content, phrase content, and word order prediction, each  ... 
arXiv:1809.06502v1 fatcat:auprpdvjsfgq7oaqumkr76n5gi

Jointly Modeling Embedding and Translation to Bridge Video and Language

Yingwei Pan, Tao Mei, Ting Yao, Houqiang Li, Yong Rui
2016 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR)  
relationship between the semantics of the entire sentence and visual content.  ...  The experiments on YouTube2Text dataset show that our proposed LSTM-E achieves to-date the best published performance in generating natural sentences: 45.3% and 31.0% in terms of BLEU@4 and METEOR, respectively  ...  sentence and video content which is measured in the visual-semantic embedding.  ... 
doi:10.1109/cvpr.2016.497 dblp:conf/cvpr/PanMYLR16 fatcat:i3zdepzpsvfqhco67ulg5hycaq

Sentence memory in aphasia

Alfonso Caramazza, Edgar B. Zurif, Howard Gardner
1978 Neuropsychologia  
Further, both groups had marked difficulties in remembering center-embedded sentences.  ...  Three types of sentences were used: active, passive and center-embedded sentences.  ...  and LYNN DAMPS for their assistance in testing the patients. We also thank RITA BERNDT, ELEANOR SAFPRAN, MYRNA SCHWARKZ for their helpful comments on an earlier version of this paper.  ... 
doi:10.1016/0028-3932(78)90001-5 pmid:748803 fatcat:zndbuzlfxrb4vjafeyagori3xm

Attentional resource and processing speed limitations during sentence processing in Parkinson's disease

Christine Lee, Murray Grossman, Jennifer Morris, Matthew B. Stern, Howard I. Hurtig
2003 Brain and Language  
We also found that PD patients are significantly slowed in their sensitivity to phonetic errors in content words embedded in object-relative center-embedded sentences.  ...  Slowed sensitivity to content words in object-relative sentences was correlated with timed executive measures of planning.  ...  Our second finding of note is that PD patients are slowed in their sensitivity to a content word located in an object-relative center-embedded sentence.  ... 
doi:10.1016/s0093-934x(03)00063-4 pmid:12744946 fatcat:kuvopgtq6ff6jeipuwotuvpxme

Irony and the dogma of force and sense

S. J. Barker, M. Popa-Wyatt
2014 Analysis  
This is the truth-conditional embedding principle: TC-Embedding: If a sentence S is embedded in a compound sentence (...S...), then the speech act performed with S is a propositional act: an act that involves  ...  However, in this paper, we restrict ourselves to figurative speech and irony in particular. We argue that irony embeds in compound sentences.  ...  One might propose that the 8 signalling itself is cancelled, due to the embedding of (1). But if that is so, how is there any ironic content at all in the embedded sentence?  ... 
doi:10.1093/analys/anu104 fatcat:zm6lazrxknhtppwyz7kococncq

Interpretable Recommendation via Attraction Modeling: Learning Multilevel Attractiveness over Multimodal Movie Contents

Liang Hu, Songlei Jian, Longbing Cao, Qingkui Chen
2018 Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence  
New contents like blogs and online videos are produced in every second in the new media age. We argue that attraction is one of the decisive factors for user selection of new contents.  ...  In particular, we design multilevel personal filters to calculate users' attractiveness on words, sentences and cast members at different levels.  ...  Denil et al. [2014] use convolutional neural networks (CNN) to transform word embedding in each sentence into the embedding for the entire sentence.  ... 
doi:10.24963/ijcai.2018/472 dblp:conf/ijcai/HuJCC18 fatcat:sawu45com5gjjlbpl2uriowgia

Content Selection in Deep Learning Models of Summarization [article]

Chris Kedzie, Kathleen McKeown, Hal Daume III
2019 arXiv   pre-print
We carry out experiments with deep learning models of summarization across the domains of news, personal stories, meetings, and medical articles in order to understand how content selection is performed  ...  At the same time, they suggest important questions for new research in summarization; namely, new forms of sentence representations or external knowledge sources are needed that are better suited to the  ...  We would also like to thank Wen Xiao for identifying an error in the oracle results for the AMI corpus, which as since been corrected.  ... 
arXiv:1810.12343v2 fatcat:varlnfq6arflbjyket7td2hhiq
« Previous Showing results 1 — 15 out of 145,858 results