RoboTALES: Learning Reasoning-Guided Robot Policies via Task-Aligned Simulated Futures
This work addresses the problem of using video generative models for robot control by improving task alignment and action-conditioning, which is important for robotics researchers working on visuomotor policy learning.
RoboTALES introduces a framework that uses hierarchical LLM-based planning and VLM-based reward feedback to guide video generative models for robot policy learning, achieving consistent improvements over baselines on RoboCasa and LIBERO10, especially in long-horizon tasks.
Pretrained video generative models are promising backbones for visuomotor control, but their imagined futures often drift from task intent and are not reliably action-conditional. As a result, these models can be difficult to use for planning or policy extraction. To address these limitations, we propose RoboTALES, a single-stage framework that learns task-aligned simulated futures and uses them to train robot policies. Our approach introduces two key innovations: (1) a hierarchical LLM-based planner that breaks complex tasks into a sequence of subgoals to guide the model's imagination; and (2) a VLM-based critic that evaluates these ``imagined'' futures and uses reward-based feedback to keep the model's internal representations focused on the goal. By anchoring the video generator in abstract reasoning, we produce temporally consistent rollouts and more coherent actions. We evaluate RoboTALES on diverse manipulation tasks from RoboCasa and LIBERO10, and show that our method consistently outperforms existing methods, especially in long-horizon tasks. Our code and models are publicly available at https://github.com/hananshafi/RoboTALES.