{"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/enhancing-sentence-embedding-with-generalized","title":"Enhancing Sentence Embedding with Generalized Pooling","arxiv_id":"1806.09828","date":"2018-06-26","proceeding":"COLING 2018 8","authors":["Qian Chen","Zhen-Hua Ling","Xiaodan Zhu"],"abstract":"Pooling is an essential component of a wide variety of sentence\nrepresentation and embedding models. This paper explores generalized pooling\nmethods to enhance sentence embedding. We propose vector-based multi-head\nattention that includes the widely used max pooling, mean pooling, and scalar\nself-attention as special cases. The model benefits from properly designed\npenalization terms to reduce redundancy in multi-head attention. We evaluate\nthe proposed model on three different tasks: natural language inference (NLI),\nauthor profiling, and sentiment classification. The experiments show that the\nproposed model achieves significant improvement over strong\nsentence-encoding-based methods, resulting in state-of-the-art performances on\nfour datasets. The proposed approach can be easily implemented for more\nproblems than we discuss in this paper.","url_abs":"http://arxiv.org/abs/1806.09828v1","url_pdf":"http://arxiv.org/pdf/1806.09828v1.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":"enhancing-sentence-embedding-with-generalized","repo_url":"https://github.com/lukecq1231/generalized-pooling","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"author-profiling","task_name":"Author Profiling"},{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"natural-language-inference","task_name":"Natural Language Inference"},{"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":"sentiment-analysis","task_name":"Sentiment Analysis"},{"task_slug":"sentiment-classification","task_name":"Sentiment Classification"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/natural-language-inference-on-snli","task":"Natural Language Inference","dataset":"SNLI","model":"600D BiLSTM with generalized pooling","rank_in_archive_order":53,"of":98,"metrics":{"% Test Accuracy":"86.6","% Train Accuracy":"94.9","Parameters":"65m"},"uses_additional_data":false},{"leaderboard":"/sota/sentiment-analysis-on-yelp-fine-grained","task":"Sentiment Analysis","dataset":"Yelp Fine-grained classification","model":"BiLSTM generalized pooling","rank_in_archive_order":10,"of":17,"metrics":{"Error":"33.45"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1806.09828","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}