{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/dr-bilstm-dependent-reading-bidirectional","title":"DR-BiLSTM: Dependent Reading Bidirectional LSTM for Natural Language Inference","arxiv_id":"1802.05577","date":"2018-02-15","proceeding":"NAACL 2018 6","authors":["Reza Ghaeini","Sadid A. Hasan","Vivek Datla","Joey Liu","Kathy Lee","Ashequl Qadir","Yuan Ling","Aaditya Prakash","Xiaoli Z. Fern","Oladimeji Farri"],"abstract":"We present a novel deep learning architecture to address the natural language\ninference (NLI) task. Existing approaches mostly rely on simple reading\nmechanisms for independent encoding of the premise and hypothesis. Instead, we\npropose a novel dependent reading bidirectional LSTM network (DR-BiLSTM) to\nefficiently model the relationship between a premise and a hypothesis during\nencoding and inference. We also introduce a sophisticated ensemble strategy to\ncombine our proposed models, which noticeably improves final predictions.\nFinally, we demonstrate how the results can be improved further with an\nadditional preprocessing step. Our evaluation shows that DR-BiLSTM obtains the\nbest single model and ensemble model results achieving the new state-of-the-art\nscores on the Stanford NLI dataset.","url_abs":"http://arxiv.org/abs/1802.05577v2","url_pdf":"http://arxiv.org/pdf/1802.05577v2.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[],"tasks":[{"task_slug":"natural-language-inference","task_name":"Natural Language Inference"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/natural-language-inference-on-snli","task":"Natural Language Inference","dataset":"SNLI","model":"450D DR-BiLSTM Ensemble","rank_in_archive_order":19,"of":98,"metrics":{"% Test Accuracy":"89.3","% Train Accuracy":"94.8","Parameters":"45m"},"uses_additional_data":false},{"leaderboard":"/sota/natural-language-inference-on-snli","task":"Natural Language Inference","dataset":"SNLI","model":"450D DR-BiLSTM","rank_in_archive_order":34,"of":98,"metrics":{"% Test Accuracy":"88.5","% Train Accuracy":"94.1","Parameters":"7.5m"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1802.05577","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}