{"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/190501976","title":"TextKD-GAN: Text Generation using KnowledgeDistillation and Generative Adversarial Networks","arxiv_id":"1905.01976","date":"2019-04-23","proceeding":null,"authors":["Md. Akmal Haidar","Mehdi Rezagholizadeh"],"abstract":"Text generation is of particular interest in many NLP applications such as\nmachine translation, language modeling, and text summarization. Generative\nadversarial networks (GANs) achieved a remarkable success in high quality image\ngeneration in computer vision,and recently, GANs have gained lots of interest\nfrom the NLP community as well. However, achieving similar success in NLP would\nbe more challenging due to the discrete nature of text. In this work, we\nintroduce a method using knowledge distillation to effectively exploit GAN\nsetup for text generation. We demonstrate how autoencoders (AEs) can be used\nfor providing a continuous representation of sentences, which is a smooth\nrepresentation that assign non-zero probabilities to more than one word. We\ndistill this representation to train the generator to synthesize similar smooth\nrepresentations. We perform a number of experiments to validate our idea using\ndifferent datasets and show that our proposed approach yields better\nperformance in terms of the BLEU score and Jensen-Shannon distance (JSD)\nmeasure compared to traditional GAN-based text generation approaches without\npre-training.","url_abs":"http://arxiv.org/abs/1905.01976v1","url_pdf":"http://arxiv.org/pdf/1905.01976v1.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":"190501976","repo_url":"https://github.com/Ankur3107/awesome-daily-blog","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"},{"task_slug":"language-modeling","task_name":"Language Modeling"},{"task_slug":"language-modelling","task_name":"Language Modelling"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"text-generation","task_name":"Text Generation"},{"task_slug":"text-summarization","task_name":"Text Summarization"},{"task_slug":"translation","task_name":"Translation"}],"methods":[{"method_slug":"knowledge-distillation","method_name":"Knowledge Distillation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}