{"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/structural-embedding-of-syntactic-trees-for","title":"Structural Embedding of Syntactic Trees for Machine Comprehension","arxiv_id":"1703.00572","date":"2017-03-02","proceeding":"EMNLP 2017 9","authors":["Rui Liu","Junjie Hu","Wei Wei","Zi Yang","Eric Nyberg"],"abstract":"Deep neural networks for machine comprehension typically utilizes only word\nor character embeddings without explicitly taking advantage of structured\nlinguistic information such as constituency trees and dependency trees. In this\npaper, we propose structural embedding of syntactic trees (SEST), an algorithm\nframework to utilize structured information and encode them into vector\nrepresentations that can boost the performance of algorithms for the machine\ncomprehension. We evaluate our approach using a state-of-the-art neural\nattention model on the SQuAD dataset. Experimental results demonstrate that our\nmodel can accurately identify the syntactic boundaries of the sentences and\nextract answers that are syntactically coherent over the baseline methods.","url_abs":"http://arxiv.org/abs/1703.00572v3","url_pdf":"http://arxiv.org/pdf/1703.00572v3.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":"question-answering","task_name":"Question Answering"},{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/question-answering-on-squad11","task":"Question Answering","dataset":"SQuAD1.1","model":"SEDT (ensemble model)","rank_in_archive_order":131,"of":213,"metrics":{"EM":"74.090","F1":"81.761"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-squad11","task":"Question Answering","dataset":"SQuAD1.1","model":"SEDT+BiDAF (ensemble)","rank_in_archive_order":135,"of":213,"metrics":{"EM":"73.723","F1":"81.530"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-squad11","task":"Question Answering","dataset":"SQuAD1.1","model":"SEDT+BiDAF (single model)","rank_in_archive_order":165,"of":213,"metrics":{"EM":"68.478","F1":"77.971"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-squad11","task":"Question Answering","dataset":"SQuAD1.1","model":"SEDT (single model)","rank_in_archive_order":168,"of":213,"metrics":{"EM":"68.163","F1":"77.527"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-squad11-dev","task":"Question Answering","dataset":"SQuAD1.1 dev","model":"SEDT-LSTM","rank_in_archive_order":40,"of":55,"metrics":{"EM":" 67.89","F1":" 77.42 "},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-squad11-dev","task":"Question Answering","dataset":"SQuAD1.1 dev","model":"SECT-LSTM","rank_in_archive_order":42,"of":55,"metrics":{"EM":"67.65","F1":"77.19"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1703.00572","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}