Neural Tree Indexers for Text Understanding. Munkhdalai, T. & Yu, H. In Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers, pages 11–21, Valencia, Spain, April, 2017. Association for Computational Linguistics. ![Neural Tree Indexers for Text Understanding [link]](https://bibbase.org/img/filetypes/link.svg "link")
Paper abstract bibtex Recurrent neural networks (RNNs) process input text sequentially and model the conditional transition between word tokens. In contrast, the advantages of recursive networks include that they explicitly model the compositionality and the recursive structure of natural language. However, the current recursive architecture is limited by its dependence on syntactic tree. In this paper, we introduce a robust syntactic parsing-independent tree structured model, Neural Tree Indexers (NTI) that provides a middle ground between the sequential RNNs and the syntactic treebased recursive models. NTI constructs a full n-ary tree by processing the input text with its node function in a bottom-up fashion. Attention mechanism can then be applied to both structure and node function. We implemented and evaluated a binary tree model of NTI, showing the model achieved the state-of-the-art performance on three different NLP tasks: natural language inference, answer sentence selection, and sentence classification, outperforming state-of-the-art recurrent and recursive neural networks.
@inproceedings{munkhdalai_neural_2017-1,
address = {Valencia, Spain},
title = {Neural {Tree} {Indexers} for {Text} {Understanding}},
url = {http://www.aclweb.org/anthology/E17-1002},
abstract = {Recurrent neural networks (RNNs) process input text sequentially and model the conditional transition between word tokens. In contrast, the advantages of recursive networks include that they explicitly model the compositionality and the recursive structure of natural language. However, the current recursive architecture is limited by its dependence on syntactic tree. In this paper, we introduce a robust syntactic parsing-independent tree structured model, Neural Tree Indexers (NTI) that provides a middle ground between the sequential RNNs and the syntactic treebased recursive models. NTI constructs a full n-ary tree by processing the input text with its node function in a bottom-up fashion. Attention mechanism can then be applied to both structure and node function. We implemented and evaluated a binary tree model of NTI, showing the model achieved the state-of-the-art performance on three different NLP tasks: natural language inference, answer sentence selection, and sentence classification, outperforming state-of-the-art recurrent and recursive neural networks.},
urldate = {2017-04-02},
booktitle = {Proceedings of the 15th {Conference} of the {European} {Chapter} of the {Association} for {Computational} {Linguistics}: {Volume} 1, {Long} {Papers}},
publisher = {Association for Computational Linguistics},
author = {Munkhdalai, Tsendsuren and Yu, Hong},
month = apr,
year = {2017},
pages = {11--21},
}
Downloads: 0
{"_id":"FrNHRLjaNXHiZRL5z","bibbaseid":"munkhdalai-yu-neuraltreeindexersfortextunderstanding-2017","author_short":["Munkhdalai, T.","Yu, H."],"bibdata":{"bibtype":"inproceedings","type":"inproceedings","address":"Valencia, Spain","title":"Neural Tree Indexers for Text Understanding","url":"http://www.aclweb.org/anthology/E17-1002","abstract":"Recurrent neural networks (RNNs) process input text sequentially and model the conditional transition between word tokens. In contrast, the advantages of recursive networks include that they explicitly model the compositionality and the recursive structure of natural language. However, the current recursive architecture is limited by its dependence on syntactic tree. In this paper, we introduce a robust syntactic parsing-independent tree structured model, Neural Tree Indexers (NTI) that provides a middle ground between the sequential RNNs and the syntactic treebased recursive models. NTI constructs a full n-ary tree by processing the input text with its node function in a bottom-up fashion. Attention mechanism can then be applied to both structure and node function. We implemented and evaluated a binary tree model of NTI, showing the model achieved the state-of-the-art performance on three different NLP tasks: natural language inference, answer sentence selection, and sentence classification, outperforming state-of-the-art recurrent and recursive neural networks.","urldate":"2017-04-02","booktitle":"Proceedings of the 15th Conference of the European Chapter of the Association for Computational Linguistics: Volume 1, Long Papers","publisher":"Association for Computational Linguistics","author":[{"propositions":[],"lastnames":["Munkhdalai"],"firstnames":["Tsendsuren"],"suffixes":[]},{"propositions":[],"lastnames":["Yu"],"firstnames":["Hong"],"suffixes":[]}],"month":"April","year":"2017","pages":"11–21","bibtex":"@inproceedings{munkhdalai_neural_2017-1,\n\taddress = {Valencia, Spain},\n\ttitle = {Neural {Tree} {Indexers} for {Text} {Understanding}},\n\turl = {http://www.aclweb.org/anthology/E17-1002},\n\tabstract = {Recurrent neural networks (RNNs) process input text sequentially and model the conditional transition between word tokens. In contrast, the advantages of recursive networks include that they explicitly model the compositionality and the recursive structure of natural language. However, the current recursive architecture is limited by its dependence on syntactic tree. In this paper, we introduce a robust syntactic parsing-independent tree structured model, Neural Tree Indexers (NTI) that provides a middle ground between the sequential RNNs and the syntactic treebased recursive models. NTI constructs a full n-ary tree by processing the input text with its node function in a bottom-up fashion. Attention mechanism can then be applied to both structure and node function. We implemented and evaluated a binary tree model of NTI, showing the model achieved the state-of-the-art performance on three different NLP tasks: natural language inference, answer sentence selection, and sentence classification, outperforming state-of-the-art recurrent and recursive neural networks.},\n\turldate = {2017-04-02},\n\tbooktitle = {Proceedings of the 15th {Conference} of the {European} {Chapter} of the {Association} for {Computational} {Linguistics}: {Volume} 1, {Long} {Papers}},\n\tpublisher = {Association for Computational Linguistics},\n\tauthor = {Munkhdalai, Tsendsuren and Yu, Hong},\n\tmonth = apr,\n\tyear = {2017},\n\tpages = {11--21},\n}\n\n","author_short":["Munkhdalai, T.","Yu, H."],"key":"munkhdalai_neural_2017-1","id":"munkhdalai_neural_2017-1","bibbaseid":"munkhdalai-yu-neuraltreeindexersfortextunderstanding-2017","role":"author","urls":{"Paper":"http://www.aclweb.org/anthology/E17-1002"},"metadata":{"authorlinks":{}}},"bibtype":"inproceedings","biburl":"http://fenway.cs.uml.edu/papers/pubs-all.bib","dataSources":["TqaA9miSB65nRfS5H"],"keywords":[],"search_terms":["neural","tree","indexers","text","understanding","munkhdalai","yu"],"title":"Neural Tree Indexers for Text Understanding","year":2017}