{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2024,3,16]],"date-time":"2024-03-16T05:04:19Z","timestamp":1710565459040},"reference-count":6,"publisher":"MIT Press - Journals","content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["TACL"],"published-print":{"date-parts":[[2018,12]]},"abstract":"<jats:p> In this work, we propose a new language modeling paradigm that has the ability to perform both prediction and moderation of information flow at multiple granularities: neural lattice language models. These models construct a lattice of possible paths through a sentence and marginalize across this lattice to calculate sequence probabilities or optimize parameters. This approach allows us to seamlessly incorporate linguistic intuitions \u2014 including polysemy and the existence of multiword lexical items \u2014 into our language model. Experiments on multiple language modeling tasks show that English neural lattice language models that utilize polysemous embeddings are able to improve perplexity by 9.95% relative to a word-level baseline, and that a Chinese model that handles multi-character tokens is able to improve perplexity by 20.94% relative to a character-level baseline. <\/jats:p>","DOI":"10.1162\/tacl_a_00036","type":"journal-article","created":{"date-parts":[[2018,12,10]],"date-time":"2018-12-10T19:32:50Z","timestamp":1544470370000},"page":"529-541","source":"Crossref","is-referenced-by-count":6,"title":["Neural Lattice Language Models"],"prefix":"10.1162","volume":"6","author":[{"given":"Jacob","family":"Buckman","sequence":"first","affiliation":[{"name":"Language Technologies Institute, Carnegie Mellon University,"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Graham","family":"Neubig","sequence":"additional","affiliation":[{"name":"Language Technologies Institute, Carnegie Mellon University,"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"281","reference":[{"issue":"3","key":"p_4","doi-asserted-by":"crossref","first-page":"241","DOI":"10.1111\/j.1467-9280.2008.02075.x","volume":"19","author":"Bannard Colin","year":"2008","journal-title":"Psychological Science"},{"key":"p_13","first-page":"239","author":"Goldwater Sharon","year":"2007","journal-title":"BUCLD 31: Proceedings of the 31st Annual Boston University Conference on Language Development, pages"},{"key":"p_15","author":"Greff Klaus","year":"2016","journal-title":"IEEE Transactions on Neural Networks and Learning Systems."},{"key":"p_16","doi-asserted-by":"publisher","DOI":"10.1162\/neco.1997.9.8.1735"},{"issue":"2","key":"p_33","doi-asserted-by":"crossref","first-page":"251","DOI":"10.1177\/0267658310382068","volume":"27","author":"Siyanova-Chanturia Anna","year":"2011","journal-title":"Second Language Research"},{"issue":"1","key":"p_39","doi-asserted-by":"crossref","first-page":"578","DOI":"10.1080\/00437956.1967.11435507","volume":"23","author":"Zgusta Ladislav","year":"1967","journal-title":"Word"}],"container-title":["Transactions of the Association for Computational Linguistics"],"original-title":[],"language":"en","link":[{"URL":"https:\/\/www.mitpressjournals.org\/doi\/pdf\/10.1162\/tacl_a_00036","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2021,3,12]],"date-time":"2021-03-12T21:38:02Z","timestamp":1615585082000},"score":1,"resource":{"primary":{"URL":"https:\/\/direct.mit.edu\/tacl\/article\/43444"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2018,12]]},"references-count":6,"alternative-id":["10.1162\/tacl_a_00036"],"URL":"https:\/\/doi.org\/10.1162\/tacl_a_00036","relation":{},"ISSN":["2307-387X"],"issn-type":[{"value":"2307-387X","type":"electronic"}],"subject":[],"published":{"date-parts":[[2018,12]]}}}