{"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/memen-multi-layer-embedding-with-memory","title":"MEMEN: Multi-layer Embedding with Memory Networks for Machine Comprehension","arxiv_id":"1707.09098","date":"2017-07-28","proceeding":null,"authors":["Boyuan Pan","Hao Li","Zhou Zhao","Bin Cao","Deng Cai","Xiaofei He"],"abstract":"Machine comprehension(MC) style question answering is a representative\nproblem in natural language processing. Previous methods rarely spend time on\nthe improvement of encoding layer, especially the embedding of syntactic\ninformation and name entity of the words, which are very crucial to the quality\nof encoding. Moreover, existing attention methods represent each query word as\na vector or use a single vector to represent the whole query sentence, neither\nof them can handle the proper weight of the key words in query sentence. In\nthis paper, we introduce a novel neural network architecture called Multi-layer\nEmbedding with Memory Network(MEMEN) for machine reading task. In the encoding\nlayer, we employ classic skip-gram model to the syntactic and semantic\ninformation of the words to train a new kind of embedding layer. We also\npropose a memory network of full-orientation matching of the query and passage\nto catch more pivotal information. Experiments show that our model has\ncompetitive results both from the perspectives of precision and efficiency in\nStanford Question Answering Dataset(SQuAD) among all published results and\nachieves the state-of-the-art results on TriviaQA dataset.","url_abs":"http://arxiv.org/abs/1707.09098v1","url_pdf":"http://arxiv.org/pdf/1707.09098v1.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"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"triviaqa","task_name":"TriviaQA"}],"methods":[{"method_slug":"memory-network","method_name":"Memory Network"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/question-answering-on-squad11","task":"Question Answering","dataset":"SQuAD1.1","model":"MEMEN  (single model)","rank_in_archive_order":91,"of":213,"metrics":{"EM":"78.234","F1":"85.344"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-squad11","task":"Question Answering","dataset":"SQuAD1.1","model":"MEMEN (single model)","rank_in_archive_order":92,"of":213,"metrics":{"EM":"78.234","F1":"85.344"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-squad11","task":"Question Answering","dataset":"SQuAD1.1","model":"MEMEN (ensemble)","rank_in_archive_order":119,"of":213,"metrics":{"EM":"75.370","F1":"82.658"},"uses_additional_data":false},{"leaderboard":"/sota/question-answering-on-triviaqa","task":"Question Answering","dataset":"TriviaQA","model":"MEMEN","rank_in_archive_order":47,"of":56,"metrics":{"EM":"43.16","F1":"46.90"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.09098","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}