{"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/neural-associative-memory-for-dual-sequence","title":"Neural Associative Memory for Dual-Sequence Modeling","arxiv_id":"1606.03864","date":"2016-06-13","proceeding":"WS 2016 8","authors":["Dirk Weissenborn"],"abstract":"Many important NLP problems can be posed as dual-sequence or\nsequence-to-sequence modeling tasks. Recent advances in building end-to-end\nneural architectures have been highly successful in solving such tasks. In this\nwork we propose a new architecture for dual-sequence modeling that is based on\nassociative memory. We derive AM-RNNs, a recurrent associative memory (AM)\nwhich augments generic recurrent neural networks (RNN). This architecture is\nextended to the Dual AM-RNN which operates on two AMs at once. Our models\nachieve very competitive results on textual entailment. A qualitative analysis\ndemonstrates that long range dependencies between source and target-sequence\ncan be bridged effectively using Dual AM-RNNs. However, an initial experiment\non auto-encoding reveals that these benefits are not exploited by the system\nwhen learning to solve sequence-to-sequence tasks which indicates that\nadditional supervision or regularization is needed.","url_abs":"http://arxiv.org/abs/1606.03864v2","url_pdf":"http://arxiv.org/pdf/1606.03864v2.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":"neural-associative-memory-for-dual-sequence","repo_url":"https://github.com/dirkweissenborn/dual_am_rnn","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"natural-language-inference","task_name":"Natural Language Inference"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}