{"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/a-hierarchical-decoding-model-for-spoken","title":"A Hierarchical Decoding Model For Spoken Language Understanding From Unaligned Data","arxiv_id":"1904.04498","date":"2019-04-09","proceeding":null,"authors":["Zijian Zhao","Su Zhu","Kai Yu"],"abstract":"Spoken language understanding (SLU) systems can be trained on two types of\nlabelled data: aligned or unaligned. Unaligned data do not require word by word\nannotation and is easier to be obtained. In the paper, we focus on spoken\nlanguage understanding from unaligned data whose annotation is a set of\nact-slot-value triples. Previous works usually focus on improve slot-value pair\nprediction and estimate dialogue act types separately, which ignores the\nhierarchical structure of the act-slot-value triples. Here, we propose a novel\nhierarchical decoding model which dynamically parses act, slot and value in a\nstructured way and employs pointer network to handle out-of-vocabulary (OOV)\nvalues. Experiments on DSTC2 dataset, a benchmark unaligned dataset, show that\nthe proposed model not only outperforms previous state-of-the-art model, but\nalso can be generalized effectively and efficiently to unseen act-slot type\npairs and OOV values.","url_abs":"http://arxiv.org/abs/1904.04498v1","url_pdf":"http://arxiv.org/pdf/1904.04498v1.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":[{"paper_slug":"a-hierarchical-decoding-model-for-spoken","repo_url":"https://github.com/simplc/WCN-BERT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"spoken-language-understanding","task_name":"Spoken Language Understanding"}],"methods":[{"method_slug":"lstm","method_name":"LSTM"},{"method_slug":"pointer-net","method_name":"Pointer Network"},{"method_slug":"sigmoid-activation","method_name":"Sigmoid Activation"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"tanh-activation","method_name":"Tanh Activation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1904.04498","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}