{"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/learning-discourse-level-diversity-for-neural","title":"Learning Discourse-level Diversity for Neural Dialog Models using Conditional Variational Autoencoders","arxiv_id":"1703.10960","date":"2017-03-31","proceeding":"ACL 2017 7","authors":["Tiancheng Zhao","Ran Zhao","Maxine Eskenazi"],"abstract":"While recent neural encoder-decoder models have shown great promise in\nmodeling open-domain conversations, they often generate dull and generic\nresponses. Unlike past work that has focused on diversifying the output of the\ndecoder at word-level to alleviate this problem, we present a novel framework\nbased on conditional variational autoencoders that captures the discourse-level\ndiversity in the encoder. Our model uses latent variables to learn a\ndistribution over potential conversational intents and generates diverse\nresponses using only greedy decoders. We have further developed a novel variant\nthat is integrated with linguistic prior knowledge for better performance.\nFinally, the training procedure is improved by introducing a bag-of-word loss.\nOur proposed models have been validated to generate significantly more diverse\nresponses than baseline approaches and exhibit competence in discourse-level\ndecision-making.","url_abs":"http://arxiv.org/abs/1703.10960v3","url_pdf":"http://arxiv.org/pdf/1703.10960v3.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":"learning-discourse-level-diversity-for-neural","repo_url":"https://github.com/snakeztc/NeuralDialog-CVAE","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}}],"tasks":[{"task_slug":"decision-making","task_name":"Decision Making"},{"task_slug":"decoder","task_name":"Decoder"},{"task_slug":"dialogue-generation","task_name":"Dialogue Generation"},{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"text-generation","task_name":"Text Generation"}],"methods":[{"method_slug":"cvae","method_name":"cVAE"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1703.10960","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}