{"ID":2831391,"CreatedAt":"2026-06-01T04:54:23.091178241Z","UpdatedAt":"2026-06-01T04:54:23.091178241Z","DeletedAt":null,"paper_url":"https://arxiv.org/abs/2512.09101","arxiv_id":"2512.09101","title":"Masked Generative Policy for Robotic Control","abstract":"We present Masked Generative Policy (MGP), a novel framework for visuomotor imitation learning. We represent actions as discrete tokens, and train a conditional masked transformer that generates tokens in parallel and then rapidly refines only low-confidence tokens. We further propose two new sampling paradigms: MGP-Short, which performs parallel masked generation with score-based refinement for Markovian tasks, and MGP-Long, which predicts full trajectories in a single pass and dynamically refines low-confidence action tokens based on new observations. With globally coherent prediction and robust adaptive execution capabilities, MGP-Long enables reliable control on complex and non-Markovian tasks that prior methods struggle with. Extensive evaluations on 150 robotic manipulation tasks spanning the Meta-World and LIBERO benchmarks show that MGP achieves both rapid inference and superior success rates compared to state-of-the-art diffusion and autoregressive policies. Specifically, MGP increases the average success rate by 9% across 150 tasks while cutting per-sequence inference time by up to 35x. It further improves the average success rate by 60% in dynamic and missing-observation environments, and solves two non-Markovian scenarios where other state-of-the-art methods fail.","short_abstract":"We present Masked Generative Policy (MGP), a novel framework for visuomotor imitation learning. We represent actions as discrete tokens, and train a conditional masked transformer that generates tokens in parallel and then rapidly refines only low-confidence tokens. We further propose two new sampling paradigms: MGP-Sh...","url_abs":"https://arxiv.org/abs/2512.09101","url_pdf":"https://arxiv.org/pdf/2512.09101v2","authors":"[\"Lipeng Zhuang\",\"Shiyu Fan\",\"Florent P. Audonnet\",\"Yingdong Ru\",\"Edmond S. L. Ho\",\"Gerardo Aragon Camarasa\",\"Paul Henderson\"]","published":"2025-12-09T20:37:40Z","proceeding":"cs.RO","tasks":"[\"cs.RO\",\"cs.AI\"]","methods":"[\"Diffusion Model\",\"Transformer\"]","has_code":false}