{"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/revisiting-adversarial-autoencoder-for","title":"Revisiting Adversarial Autoencoder for Unsupervised Word Translation with Cycle Consistency and Improved Training","arxiv_id":"1904.04116","date":"2019-04-04","proceeding":"NAACL 2019 6","authors":["Tasnim Mohiuddin","Shafiq Joty"],"abstract":"Adversarial training has shown impressive success in learning bilingual\ndictionary without any parallel data by mapping monolingual embeddings to a\nshared space. However, recent work has shown superior performance for\nnon-adversarial methods in more challenging language pairs. In this work, we\nrevisit adversarial autoencoder for unsupervised word translation and propose\ntwo novel extensions to it that yield more stable training and improved\nresults. Our method includes regularization terms to enforce cycle consistency\nand input reconstruction, and puts the target encoders as an adversary against\nthe corresponding discriminator. Extensive experimentations with European,\nnon-European and low-resource languages show that our method is more robust and\nachieves better performance than recently proposed adversarial and\nnon-adversarial approaches.","url_abs":"http://arxiv.org/abs/1904.04116v1","url_pdf":"http://arxiv.org/pdf/1904.04116v1.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":"revisiting-adversarial-autoencoder-for","repo_url":"https://github.com/taasnim/unsup-word-translation","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"translation","task_name":"Translation"},{"task_slug":"word-translation","task_name":"Word Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1904.04116","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}