{"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/hierarchical-attention-what-really-counts-in","title":"Hierarchical Attention: What Really Counts in Various NLP Tasks","arxiv_id":"1808.03728","date":"2018-08-10","proceeding":null,"authors":["Zehao Dou","Zhihua Zhang"],"abstract":"Attention mechanisms in sequence to sequence models have shown great ability\nand wonderful performance in various natural language processing (NLP) tasks,\nsuch as sentence embedding, text generation, machine translation, machine\nreading comprehension, etc. Unfortunately, existing attention mechanisms only\nlearn either high-level or low-level features. In this paper, we think that the\nlack of hierarchical mechanisms is a bottleneck in improving the performance of\nthe attention mechanisms, and propose a novel Hierarchical Attention Mechanism\n(Ham) based on the weighted sum of different layers of a multi-level attention.\nHam achieves a state-of-the-art BLEU score of 0.26 on Chinese poem generation\ntask and a nearly 6.5% averaged improvement compared with the existing machine\nreading comprehension models such as BIDAF and Match-LSTM. Furthermore, our\nexperiments and theorems reveal that Ham has greater generalization and\nrepresentation ability than existing attention mechanisms.","url_abs":"http://arxiv.org/abs/1808.03728v1","url_pdf":"http://arxiv.org/pdf/1808.03728v1.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":"hierarchical-attention-what-really-counts-in","repo_url":"https://github.com/Disiok/poetry-seq2seq","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"machine-reading-comprehension","task_name":"Machine Reading Comprehension"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"reading-comprehension","task_name":"Reading Comprehension"},{"task_slug":"sentence","task_name":"Sentence"},{"task_slug":"sentence-embedding","task_name":"Sentence Embedding"},{"task_slug":"sentence-embedding-1","task_name":"Sentence-Embedding"},{"task_slug":"text-generation","task_name":"Text Generation"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}