Compositional Distributional Semantic Models (CDSMs) model the meaning of phrases and sentences in vector space. They have been predominantly evaluated on limited, artificial tasks such as semantic sentence similarity on hand-constructed datasets. This paper argues for lexical substitution as a means to evaluate CDSMs. Lexical substitution is a more natural task, enables us to evaluate meaning composition at the level of individual words, and provides a common ground to compare CDSMs with dedicated lexical substitution models. We create a lexical substitution dataset for CDSM evaluation from an English-language corpus with manual “all-words” lexical substitution annotation. Our experiments indicate that the Practical Lexical Function CDSM outperforms simple component-wise CDSMs and performs on par with the context2vec lexical substitution model using the same context.
%0 Conference Paper
%1 buljan18:_lexic_subst_evaluat_compos_distr_model
%A Buljan, Maja
%A Padó, Sebastian
%A Snajder, Jan
%B Proceedings of NAACL
%C New Orleans, LA
%D 2018
%K conference myown
%P 206-211
%T Lexical Substitution for Evaluating Compositional Distributional Models
%U http://aclweb.org/anthology/N18-2033.pdf
%X Compositional Distributional Semantic Models (CDSMs) model the meaning of phrases and sentences in vector space. They have been predominantly evaluated on limited, artificial tasks such as semantic sentence similarity on hand-constructed datasets. This paper argues for lexical substitution as a means to evaluate CDSMs. Lexical substitution is a more natural task, enables us to evaluate meaning composition at the level of individual words, and provides a common ground to compare CDSMs with dedicated lexical substitution models. We create a lexical substitution dataset for CDSM evaluation from an English-language corpus with manual “all-words” lexical substitution annotation. Our experiments indicate that the Practical Lexical Function CDSM outperforms simple component-wise CDSMs and performs on par with the context2vec lexical substitution model using the same context.
@inproceedings{buljan18:_lexic_subst_evaluat_compos_distr_model,
abstract = {Compositional Distributional Semantic Models (CDSMs) model the meaning of phrases and sentences in vector space. They have been predominantly evaluated on limited, artificial tasks such as semantic sentence similarity on hand-constructed datasets. This paper argues for lexical substitution as a means to evaluate CDSMs. Lexical substitution is a more natural task, enables us to evaluate meaning composition at the level of individual words, and provides a common ground to compare CDSMs with dedicated lexical substitution models. We create a lexical substitution dataset for CDSM evaluation from an English-language corpus with manual “all-words” lexical substitution annotation. Our experiments indicate that the Practical Lexical Function CDSM outperforms simple component-wise CDSMs and performs on par with the context2vec lexical substitution model using the same context.},
added-at = {2018-03-01T11:20:32.000+0100},
address = {New Orleans, LA},
author = {Buljan, Maja and Padó, Sebastian and \v{S}najder, Jan},
biburl = {https://puma.ub.uni-stuttgart.de/bibtex/2a3a6d336afa314fd11e25cd873dcaf58/sp},
booktitle = {Proceedings of NAACL},
interhash = {fe11391937f5ef1ecf72edebf4ccffaa},
intrahash = {a3a6d336afa314fd11e25cd873dcaf58},
keywords = {conference myown},
pages = {206-211},
timestamp = {2018-06-07T08:19:59.000+0200},
title = {Lexical Substitution for Evaluating Compositional Distributional Models},
url = {http://aclweb.org/anthology/N18-2033.pdf},
year = 2018
}
