{"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/second-order-word-embeddings-from-nearest","title":"Second-Order Word Embeddings from Nearest Neighbor Topological Features","arxiv_id":"1705.08488","date":"2017-05-23","proceeding":null,"authors":["Denis Newman-Griffis","Eric Fosler-Lussier"],"abstract":"We introduce second-order vector representations of words, induced from\nnearest neighborhood topological features in pre-trained contextual word\nembeddings. We then analyze the effects of using second-order embeddings as\ninput features in two deep natural language processing models, for named entity\nrecognition and recognizing textual entailment, as well as a linear model for\nparaphrase recognition. Surprisingly, we find that nearest neighbor information\nalone is sufficient to capture most of the performance benefits derived from\nusing pre-trained word embeddings. Furthermore, second-order embeddings are\nable to handle highly heterogeneous data better than first-order\nrepresentations, though at the cost of some specificity. Additionally,\naugmenting contextual embeddings with second-order information further improves\nmodel performance in some cases. Due to variance in the random initializations\nof word embeddings, utilizing nearest neighbor features from multiple\nfirst-order embedding samples can also contribute to downstream performance\ngains. Finally, we identify intriguing characteristics of second-order\nembedding spaces for further research, including much higher density and\ndifferent semantic interpretations of cosine similarity.","url_abs":"http://arxiv.org/abs/1705.08488v1","url_pdf":"http://arxiv.org/pdf/1705.08488v1.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":"second-order-word-embeddings-from-nearest","repo_url":"https://github.com/drgriffis/knn-embedding","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[{"task_slug":"named-entity-recognition-1","task_name":"Named Entity Recognition"},{"task_slug":"named-entity-recognition-ner","task_name":"Named Entity Recognition (NER)"},{"task_slug":"natural-language-inference","task_name":"Natural Language Inference"},{"task_slug":"specificity","task_name":"Specificity"},{"task_slug":"word-embeddings","task_name":"Word Embeddings"},{"task_slug":"named-entity-recognition","task_name":"named-entity-recognition"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}