论文/arXiv 2610.05483·24 天前

FLEX-WAM：用于长时域想象与规划的灵活块因果世界-动作模型

FLEX-WAM用块因果可缓存世界-动作模型解决长时域高效稳定想象与规划，并在仿真、真实数据和机器人上验证。

AI 整理

• 仿真与训练
• 数据与采集
• 端侧部署

原文链接：FLEX-WAM: Flexible Block-Causal World-Action Models for Long-Horizon Imagination and Planning（arXiv 2610.05483 · 发布日期：2026-10-04）

FLEX-WAM: Flexible Block-Causal World-Action Models for Long-Horizon Imagination and Planning

FLEX-WAM用块因果可缓存世界-动作模型解决长时域高效稳定想象与规划，并在仿真、真实数据和机器人上验证。

核心要点

• 现有联合视频-动作模型计算开销大、固定时域，难适配流式推断和稳定长时域开环滚动。
• 模型支持可变长度上下文、非因果预测时域，以及逐帧或逐块无限自回归生成。
• 采用块因果KV缓存架构，结合轴向注意力和分块diffusion forcing提升部署效率。
• 在仿真和真实数据上提升多步预测质量与延迟，并在MCTS中完成长时域想象规划任务。

验证范围与局限：验证范围包括仿真和真实世界数据集、PushT与OGBench Puzzle-4x4想象规划任务，以及基于OpenArm的双臂机器人。

英文摘要 (Abstract)

World--action models (WAMs) promise a unified model that predicts action-conditioned futures, generates feasible actions, and supports planning in imagination. However, existing joint video--action models often use computationally heavy, fixed-horizon backbones ill-suited to streaming inference and stable long-horizon open-loop rollouts. We introduce FLEX-WAM, a Flexible and Efficient Block-Causal World--Action Model for unified simulation and policy inference. FLEX-WAM supports variable-length contexts and non-causal prediction horizons, as well as infinite autoregressive generation frame by frame or block by block. Its block-causal, KV-cacheable architecture combines axial attention and blockwise diffusion forcing to enable efficient real-time rollout and deployment-time latency--throughput tradeoffs without retraining. Joint training can nevertheless produce plausible futures that weakly respond to commanded actions. We address this failure mode by balancing state and action flow-matching gradient contributions across the state--action diffusion-noise grid and regulating world-model sampling using Forward-Dynamics (FD) elasticity, an efficient training-time proxy for action responsiveness. Across simulated and real-world datasets, FLEX-WAM achieves superior multi-step prediction quality and latency while producing stable joint state--action rollouts for thousands of steps. As a joint action proposer and simulator within MCTS, it solves long-horizon PushT and all five OGBench Puzzle-4x4 tasks entirely in imagination. On a bimanual OpenArm-based robot, a single checkpoint jointly serves as a play policy and expected-outcome predictor, enabling real-time identification and collection of model--reality mismatches for future self-improvement.

作者：Rooholla Khorrambakht, Joseph Amigo, Félix Lebel, Leon Seetoo, Jean Ponce, Zhenzhen Li, Ludovic Righetti

来源

本页事实对应的公开页面，以原文为准。

1. arXiv 2610.05483：FLEX-WAM: Flexible Block-Causal World-Action Models for Long-Horizon Imagination and Planning访问 2026-10-21