{"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/cross-temporal-recurrent-networks-for-ranking","title":"Cross Temporal Recurrent Networks for Ranking Question Answer Pairs","arxiv_id":"1711.07656","date":"2017-11-21","proceeding":null,"authors":["Yi Tay","Luu Anh Tuan","Siu Cheung Hui"],"abstract":"Temporal gates play a significant role in modern recurrent-based neural\nencoders, enabling fine-grained control over recursive compositional operations\nover time. In recurrent models such as the long short-term memory (LSTM),\ntemporal gates control the amount of information retained or discarded over\ntime, not only playing an important role in influencing the learned\nrepresentations but also serving as a protection against vanishing gradients.\nThis paper explores the idea of learning temporal gates for sequence pairs\n(question and answer), jointly influencing the learned representations in a\npairwise manner. In our approach, temporal gates are learned via 1D\nconvolutional layers and then subsequently cross applied across question and\nanswer for joint learning. Empirically, we show that this conceptually simple\nsharing of temporal gates can lead to competitive performance across multiple\nbenchmarks. Intuitively, what our network achieves can be interpreted as\nlearning representations of question and answer pairs that are aware of what\neach other is remembering or forgetting, i.e., pairwise temporal gating. Via\nextensive experiments, we show that our proposed model achieves\nstate-of-the-art performance on two community-based QA datasets and competitive\nperformance on one factoid-based QA dataset.","url_abs":"http://arxiv.org/abs/1711.07656v1","url_pdf":"http://arxiv.org/pdf/1711.07656v1.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":"cross-temporal-recurrent-networks-for-ranking","repo_url":"https://github.com/vanzytay/YahooQA_Splits","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1711.07656","atlas_url":"https://app.syntology.ai/?focus=1711.07656","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}