{"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/language-generation-with-recurrent-generative","title":"Language Generation with Recurrent Generative Adversarial Networks without Pre-training","arxiv_id":"1706.01399","date":"2017-06-05","proceeding":null,"authors":["Ofir Press","Amir Bar","Ben Bogin","Jonathan Berant","Lior Wolf"],"abstract":"Generative Adversarial Networks (GANs) have shown great promise recently in\nimage generation. Training GANs for language generation has proven to be more\ndifficult, because of the non-differentiable nature of generating text with\nrecurrent neural networks. Consequently, past work has either resorted to\npre-training with maximum-likelihood or used convolutional networks for\ngeneration. In this work, we show that recurrent neural networks can be trained\nto generate text with GANs from scratch using curriculum learning, by slowly\nteaching the model to generate sequences of increasing and variable length. We\nempirically show that our approach vastly improves the quality of generated\nsequences compared to a convolutional baseline.","url_abs":"http://arxiv.org/abs/1706.01399v3","url_pdf":"http://arxiv.org/pdf/1706.01399v3.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":"language-generation-with-recurrent-generative","repo_url":"https://github.com/amirbar/rnn.wgan","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"language-generation-with-recurrent-generative","repo_url":"https://github.com/GuyTevet/rnn-gan-eval","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"language-generation-with-recurrent-generative","repo_url":"https://github.com/valko073/LyricsGANs","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"text-generation","task_name":"Text Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1706.01399","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}