{"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/efficient-vector-representation-for-documents","title":"Efficient Vector Representation for Documents through Corruption","arxiv_id":"1707.02377","date":"2017-07-08","proceeding":null,"authors":["Minmin Chen"],"abstract":"We present an efficient document representation learning framework, Document\nVector through Corruption (Doc2VecC). Doc2VecC represents each document as a\nsimple average of word embeddings. It ensures a representation generated as\nsuch captures the semantic meanings of the document during learning. A\ncorruption model is included, which introduces a data-dependent regularization\nthat favors informative or rare words while forcing the embeddings of common\nand non-discriminative ones to be close to zero. Doc2VecC produces\nsignificantly better word embeddings than Word2Vec. We compare Doc2VecC with\nseveral state-of-the-art document representation learning algorithms. The\nsimple model architecture introduced by Doc2VecC matches or out-performs the\nstate-of-the-art in generating high-quality document representations for\nsentiment analysis, document classification as well as semantic relatedness\ntasks. The simplicity of the model enables training on billions of words per\nhour on a single machine. At the same time, the model is very efficient in\ngenerating representations of unseen documents at test time.","url_abs":"http://arxiv.org/abs/1707.02377v1","url_pdf":"http://arxiv.org/pdf/1707.02377v1.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":"efficient-vector-representation-for-documents","repo_url":"https://github.com/mchen24/iclr2017","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":null}],"tasks":[{"task_slug":"document-classification","task_name":"Document Classification"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"sentiment-analysis","task_name":"Sentiment Analysis"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/semantic-similarity-on-sick","task":"Semantic Similarity","dataset":"SICK","model":"Doc2VecC","rank_in_archive_order":5,"of":5,"metrics":{"MSE":"0.3053","Pearson Correlation":"0.8381","Spearman Correlation":"0.7621"},"uses_additional_data":false},{"leaderboard":"/sota/sentiment-analysis-on-imdb","task":"Sentiment Analysis","dataset":"IMDb","model":"Doc2VecC","rank_in_archive_order":41,"of":49,"metrics":{"Accuracy":"88.3"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1707.02377","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}