{"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/learning-to-rank-question-answer-pairs-with","title":"Learning to Rank Question Answer Pairs with Holographic Dual LSTM Architecture","arxiv_id":"1707.06372","date":"2017-07-20","proceeding":null,"authors":["Tay Yi","Phan Minh C.","Tuan Luu Anh","Hui Siu Cheung"],"abstract":"We describe a new deep learning architecture for learning to rank question\nanswer pairs. Our approach extends the long short-term memory (LSTM) network\nwith holographic composition to model the relationship between question and\nanswer representations. As opposed to the neural tensor layer that has been\nadopted recently, the holographic composition provides the benefits of scalable\nand rich representational learning approach without incurring huge parameter\ncosts. Overall, we present Holographic Dual LSTM (HD-LSTM), a unified\narchitecture for both deep sentence modeling and semantic matching.\nEssentially, our model is trained end-to-end whereby the parameters of the LSTM\nare optimized in a way that best explains the correlation between question and\nanswer representations. In addition, our proposed deep learning architecture\nrequires no extensive feature engineering. Via extensive experiments, we show\nthat HD-LSTM outperforms many other neural architectures on two popular\nbenchmark QA datasets. Empirical studies confirm the effectiveness of\nholographic composition over the neural tensor layer.","url_abs":"http://arxiv.org/abs/1707.06372v1","url_pdf":"http://arxiv.org/pdf/1707.06372v1.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":"learning-to-rank-question-answer-pairs-with","repo_url":"https://github.com/lezzhov/learning_to_rank","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[{"task_slug":"feature-engineering","task_name":"Feature Engineering"},{"task_slug":"learning-to-rank","task_name":"Learning-To-Rank"},{"task_slug":"sentence","task_name":"Sentence"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1707.06372","atlas_url":"https://app.syntology.ai/?focus=1707.06372","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}