← 研踪 PaperTrailarXiv:2611.17626 · v1
Joint World Action Models through Differentiable Trajectory Optimization
首次提交 2026-11-13 08:54
版本更新 2026-11-13 08:54
雷达观测 2026-11-17 04:57
摘要
Joint world action models (WAMs) generate future actions and states together, modeling world-action trajectories rather than actions alone. However, trajectory imitation does not ensure that these trajectories are task-effective or consistent with dynamics. We present \textbf{world action trajectory optimization (WATO)}, which optimizes generated actions and future states jointly for control. WATO uses both as references in a dynamics-constrained trajectory optimization problem and differentiates the task loss through the optimized trajectory to train the generator. This makes generated trajectories task-aware and dynamics-consistent while using trajectory optimization only during training. At deployment, WATO directly generates candidate trajectories and ranks them with a world model, without iterative trajectory optimization. Across visual control environments, WATO improves average success from for the imitation joint WAM to , outperforming strong model-based, goal-conditioned RL, and generative control baselines. Compared with cross-entropy method (CEM), WATO is on average faster and uses fewer world model evaluations at deployment.
作者备注:29 pages, 6 figures, 14 tables
观测到的版本
相邻记录可能跳过中间版本;首次发现 v2 及以上也会标为版本更新。
v1 · 2026-11-13 08:54 · 最新观测版本
Joint World Action Models through Differentiable Trajectory Optimization
Joint world action models (WAMs) generate future actions and states together, modeling world-action trajectories rather than actions alone. However, trajectory imitation does not ensure that these trajectories are task-effective or consistent with dynamics. We present \textbf{world action trajectory optimization (WATO)}, which optimizes generated actions and future states jointly for control. WATO uses both as references in a dynamics-constrained trajectory optimization problem and differentiates the task loss through the optimized trajectory to train the generator. This makes generated trajectories task-aware and dynamics-consistent while using trajectory optimization only during training. At deployment, WATO directly generates candidate trajectories and ranks them with a world model, without iterative trajectory optimization. Across visual control environments, WATO improves average success from for the imitation joint WAM to , outperforming strong model-based, goal-conditioned RL, and generative control baselines. Compared with cross-entropy method (CEM), WATO is on average faster and uses fewer world model evaluations at deployment.
查看此版本解读 →查看此版本原文 ↗