{"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/generating-informative-and-diverse","title":"Generating Informative and Diverse Conversational Responses via Adversarial Information Maximization","arxiv_id":"1809.05972","date":"2018-09-16","proceeding":"NeurIPS 2018 12","authors":["Yizhe Zhang","Michel Galley","Jianfeng Gao","Zhe Gan","Xiujun Li","Chris Brockett","Bill Dolan"],"abstract":"Responses generated by neural conversational models tend to lack\ninformativeness and diversity. We present Adversarial Information Maximization\n(AIM), an adversarial learning strategy that addresses these two related but\ndistinct problems. To foster response diversity, we leverage adversarial\ntraining that allows distributional matching of synthetic and real responses.\nTo improve informativeness, our framework explicitly optimizes a variational\nlower bound on pairwise mutual information between query and response.\nEmpirical results from automatic and human evaluations demonstrate that our\nmethods significantly boost informativeness and diversity.","url_abs":"http://arxiv.org/abs/1809.05972v5","url_pdf":"http://arxiv.org/pdf/1809.05972v5.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":"generating-informative-and-diverse","repo_url":"https://github.com/GraphGrailAi/DialoGPT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"generating-informative-and-diverse","repo_url":"https://github.com/dreasysnail/converse_GAN","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok"}},{"paper_slug":"generating-informative-and-diverse","repo_url":"https://github.com/souvikdgp16/dialo_gpt_daily_dialog","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"generating-informative-and-diverse","repo_url":"https://github.com/sseol11/DialoGPT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"conversational-response-generation","task_name":"Conversational Response Generation"},{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"informativeness","task_name":"Informativeness"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1809.05972","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}