{"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/unified-pragmatic-models-for-generating-and","title":"Unified Pragmatic Models for Generating and Following Instructions","arxiv_id":"1711.04987","date":"2017-11-14","proceeding":"NAACL 2018 6","authors":["Daniel Fried","Jacob Andreas","Dan Klein"],"abstract":"We show that explicit pragmatic inference aids in correctly generating and\nfollowing natural language instructions for complex, sequential tasks. Our\npragmatics-enabled models reason about why speakers produce certain\ninstructions, and about how listeners will react upon hearing them. Like\nprevious pragmatic models, we use learned base listener and speaker models to\nbuild a pragmatic speaker that uses the base listener to simulate the\ninterpretation of candidate descriptions, and a pragmatic listener that reasons\ncounterfactually about alternative descriptions. We extend these models to\ntasks with sequential structure. Evaluation of language generation and\ninterpretation shows that pragmatic inference improves state-of-the-art\nlistener models (at correctly interpreting human instructions) and speaker\nmodels (at producing instructions correctly interpreted by humans) in diverse\nsettings.","url_abs":"http://arxiv.org/abs/1711.04987v3","url_pdf":"http://arxiv.org/pdf/1711.04987v3.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":"unified-pragmatic-models-for-generating-and","repo_url":"https://github.com/dpfried/pragmatic-instructions","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"none","reach":{"status":"ok"}}],"tasks":[{"task_slug":"text-generation","task_name":"Text Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":"https://app.syntology.ai/?focus=1711.04987","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}