Filters








7 Hits in 1.1 sec

Word2Sense: Sparse Interpretable Word Embeddings

Abhishek Panigrahi, Harsha Vardhan Simhadri, Chiranjib Bhattacharyya
<span title="">2019</span> <i title="Association for Computational Linguistics"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/5n6volmnonf5tn6xputi5f2t3e" style="color: black;">Proceedings of the 57th Annual Meeting of the Association for Computational Linguistics</a> </i> &nbsp;
Word2Sense embeddings are at least as sparse and fast to compute as prior art.  ...  We present an unsupervised method to generate Word2Sense word embeddings that are interpretable -each dimension of the embedding space corresponds to a fine-grained sense, and the non-negative value of  ...  (Subramanian et al., 2018) use a k-sparse denoising autoencoder to produce sparse non-negative high dimensional projection of word embeddings, which they called SParse Interpretable Neural Embeddings  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/p19-1570">doi:10.18653/v1/p19-1570</a> <a target="_blank" rel="external noopener" href="https://dblp.org/rec/conf/acl/PanigrahiSB19.html">dblp:conf/acl/PanigrahiSB19</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/zho5rtdoafeefe6uwuxlwty5ii">fatcat:zho5rtdoafeefe6uwuxlwty5ii</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200508035243/https://www.aclweb.org/anthology/P19-1570.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/a0/1f/a01f3039dd2ef75b3db08e4cedd0fcf7139f465c.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/p19-1570"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Word Equations: Inherently Interpretable Sparse Word Embeddingsthrough Sparse Coding [article]

Adly Templeton
<span title="2021-09-27">2021</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
In order to create more interpretable word embeddings, we transform pretrained dense word embeddings into sparse embeddings.  ...  We construct these embeddings through sparse coding, where each vector in the basis set is itself a word embedding.  ...  Word2sense: Sparse interpretable word embeddings. In Proceed- ings of the 57th Annual Meeting of the Association for Computational Linguistics, pages 5692-5705.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.13847v3">arXiv:2004.13847v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/betalo7gzjhwvlpxisjbyhueeu">fatcat:betalo7gzjhwvlpxisjbyhueeu</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20210915011806/https://arxiv.org/pdf/2004.13847v2.pdf" title="fulltext PDF download [not primary version]" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <span style="color: #f43e3e;">&#10033;</span> <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/10/9c/109ccd4748e1c9ee7b750a42cb8e31340fee28c1.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2004.13847v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Word Equations: Inherently Interpretable Sparse Word Embeddings through Sparse Coding

Adly Templeton
<span title="">2021</span> <i title="Association for Computational Linguistics"> Proceedings of the Fourth BlackboxNLP Workshop on Analyzing and Interpreting Neural Networks for NLP </i> &nbsp; <span class="release-stage">unpublished</span>
In order to create more interpretable word embeddings, we transform pretrained dense word embeddings into sparse embeddings.  ...  We construct these embeddings through sparse coding, where each vector in the basis set is itself a word embedding.  ...  ., 2019) proposed Word2Sense, a generative approach that models each dimension as a 'sense' and word embeddings as a sparse probability distribution over the senses.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/2021.blackboxnlp-1.12">doi:10.18653/v1/2021.blackboxnlp-1.12</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/aq2wh2hm7jbbvkxxbsq2swigju">fatcat:aq2wh2hm7jbbvkxxbsq2swigju</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20220202020512/https://aclanthology.org/2021.blackboxnlp-1.12.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/e7/95/e795dc653dbbec3aeae75c080c2fa3feb3871784.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.18653/v1/2021.blackboxnlp-1.12"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>

Imparting Interpretability to Word Embeddings while Preserving Semantic Structure [article]

Lutfi Kerem Senel, Ihsan Utlu, Furkan Şahinuç, Haldun M. Ozaktas, Aykut Koç
<span title="2020-02-14">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Therefore, we impart interpretability to the word embedding by assigning meaning to its vector dimensions.  ...  These tests show that the interpretability-imparted word embeddings that are obtained by the proposed framework do not sacrifice performances in common benchmark tests.  ...  They also proposed a method to learn sparse interpretable word embedding, called Word2Sense, based on the obtained sense distributions.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1807.07279v3">arXiv:1807.07279v3</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/r4lf34zjajdidhiqbi274q46w4">fatcat:r4lf34zjajdidhiqbi274q46w4</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200321093300/https://arxiv.org/pdf/1807.07279v3.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/1807.07279v3" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

The POLAR Framework: Polar Opposites Enable Interpretability of Pre-Trained Word Embeddings [article]

Binny Mathew, Sandipan Sikdar, Florian Lemmerich, Markus Strohmaier
<span title="2020-01-28">2020</span> <i > arXiv </i> &nbsp; <span class="release-stage" >pre-print</span>
Our work is relevant for researchers and engineers interested in interpreting pre-trained word embeddings.  ...  We introduce POLAR - a framework that adds interpretability to pre-trained word embeddings via the adoption of semantic differentials.  ...  In that regard, Murphy et al. proposed to use a Non-Negative Sparse Embedding (NNSE) in order to to obtain sparse and interpretable word embeddings [23] . Fyshe et al.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2001.09876v2">arXiv:2001.09876v2</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/htecohfqzvfxrjueg32eftmpde">fatcat:htecohfqzvfxrjueg32eftmpde</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20200930004250/https://arxiv.org/pdf/2001.09876v2.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/25/9e/259e58c2555742982358a8df3dcf33c0843a5a70.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener" href="https://arxiv.org/abs/2001.09876v2" title="arxiv.org access"> <button class="ui compact blue labeled icon button serp-button"> <i class="file alternate outline icon"></i> arxiv.org </button> </a>

Nasari : Integrating explicit knowledge and corpus statistics for a multilingual representation of concepts and entities

José Camacho-Collados, Mohammad Taher Pilehvar, Roberto Navigli
<span title="">2016</span> <i title="Elsevier BV"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/weoplee4x5anpi62cco5v4higa" style="color: black;">Artificial Intelligence</a> </i> &nbsp;
Firstly, we took the pre-trained word embeddings of Word2Vec[82] 17 , the same used for our Nasari embed system (see Section 4.2).  ...  We propose a new flexible way to get continuous embedded vector representations, with the added benefit of obtaining a semantic space shared by BabelNet synsets, words and texts (Section 3.3). 3.  ...  The dimensions are not interpretable in the embedded vectors.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1016/j.artint.2016.07.005">doi:10.1016/j.artint.2016.07.005</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/qemh2zkstbbjlklzoeww2thzea">fatcat:qemh2zkstbbjlklzoeww2thzea</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20190429142653/http://orca.cf.ac.uk/113132/1/NASARI_AIJ.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/f6/af/f6af219aa12ff08b65d81e06f68c87eb86e6c198.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.1016/j.artint.2016.07.005"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> elsevier.com </button> </a>

Soft Cardinality in Semantic Text Processing: Experience of the SemEval International Competitions

Sergio Jimenez, Fabio A. Gonzalez, Alexander Gelbukh
<span title="2015-01-31">2015</span> <i title="Centro de Innovacion y Desarrollo Tecnologico en Computo"> <a target="_blank" rel="noopener" href="https://fatcat.wiki/container/msuxglkxsfc65p2nc6skdnp74i" style="color: black;">POLIBITS Research Journal on Computer Science and Computer Engineering With Applications</a> </i> &nbsp;
Recently, neural word embedding [28] , [29] has become the state-of-the-art for semantic lexical similarity.  ...  Removing stopwords may be interpreted as a binary weighting for the words in a text, i.e., assigning 1 for non-stopwords and 0 otherwise.  ... 
<span class="external-identifiers"> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.17562/pb-51-9">doi:10.17562/pb-51-9</a> <a target="_blank" rel="external noopener" href="https://fatcat.wiki/release/ugtspimajrbqbasm2yc3wc4hea">fatcat:ugtspimajrbqbasm2yc3wc4hea</a> </span>
<a target="_blank" rel="noopener" href="https://web.archive.org/web/20170814125626/http://www.scielo.org.mx/pdf/poli/n51/n51a10.pdf" title="fulltext PDF download" data-goatcounter-click="serp-fulltext" data-goatcounter-title="serp-fulltext"> <button class="ui simple right pointing dropdown compact black labeled icon button serp-button"> <i class="icon ia-icon"></i> Web Archive [PDF] <div class="menu fulltext-thumbnail"> <img src="https://blobs.fatcat.wiki/thumbnail/pdf/e0/f4/e0f42e0b573cd3c46c888d28c385d93ac96fd90d.180px.jpg" alt="fulltext thumbnail" loading="lazy"> </div> </button> </a> <a target="_blank" rel="external noopener noreferrer" href="https://doi.org/10.17562/pb-51-9"> <button class="ui left aligned compact blue labeled icon button serp-button"> <i class="external alternate icon"></i> Publisher / doi.org </button> </a>