{"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/plan-attend-generate-planning-for-sequence-to","title":"Plan, Attend, Generate: Planning for Sequence-to-Sequence Models","arxiv_id":"1711.10462","date":"2017-11-28","proceeding":"NeurIPS 2017 12","authors":["Francis Dutil","Caglar Gulcehre","Adam Trischler","Yoshua Bengio"],"abstract":"We investigate the integration of a planning mechanism into\nsequence-to-sequence models using attention. We develop a model which can plan\nahead in the future when it computes its alignments between input and output\nsequences, constructing a matrix of proposed future alignments and a commitment\nvector that governs whether to follow or recompute the plan. This mechanism is\ninspired by the recently proposed strategic attentive reader and writer (STRAW)\nmodel for Reinforcement Learning. Our proposed model is end-to-end trainable\nusing primarily differentiable operations. We show that it outperforms a strong\nbaseline on character-level translation tasks from WMT'15, the algorithmic task\nof finding Eulerian circuits of graphs, and question generation from the text.\nOur analysis demonstrates that the model computes qualitatively intuitive\nalignments, converges faster than the baselines, and achieves superior\nperformance with fewer parameters.","url_abs":"http://arxiv.org/abs/1711.10462v1","url_pdf":"http://arxiv.org/pdf/1711.10462v1.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":"plan-attend-generate-planning-for-sequence-to","repo_url":"https://github.com/nyu-dl/dl4mt-cdec","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"none","reach":{"status":"ok","spdx":"BSD-3-Clause"}}],"tasks":[{"task_slug":"question-generation","task_name":"Question Generation"},{"task_slug":"question-generation","task_name":"Question-Generation"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"translation","task_name":"Translation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1711.10462","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}