MIND-V: Hierarchical World Model for Long-Horizon Robotic Manipulation with RL-based Physical Alignment

TL;DR

MIND-V is a hierarchical world model that synthesizes long-horizon, physically plausible videos of robotic manipulation by integrating high-level reasoning, domain-invariant representations, and physics-aware reinforcement learning.

Contribution

The paper introduces MIND-V, a novel hierarchical world model that combines cognitive-inspired components and physics-based training for scalable long-horizon robotic simulation.

Findings

Achieves state-of-the-art performance in long-horizon simulation tasks.

Enhances policy learning through realistic, physics-consistent data.

Demonstrates robustness with staged visual future rollouts.

Abstract

Scalable embodied intelligence is constrained by the scarcity of diverse, long-horizon robotic manipulation data. Existing video world models in this domain are limited to synthesizing short clips of simple actions and often rely on manually defined trajectories. To this end, we introduce MIND-V, a cognitive hierarchical world model designed to synthesize physically plausible and logically coherent videos of long-horizon robotic manipulation. Inspired by cognitive science, MIND-V bridges high-level reasoning with pixel-level synthesis through three core components: a Semantic Reasoning Hub (SRH) that leverages a pre-trained vision-language model for task planning; a Behavioral Semantic Bridge (BSB) that translates abstract instructions into domain-invariant representations; and a Motor Video Generator (MVG) for conditional video rendering. MIND-V employs Staged Visual Future Rollouts, a test-time optimization strategy to enhance long-horizon robustness. To enforce adherence to physical laws, we introduce a GRPO reinforcement learning post-training phase guided by a novel Physical Foresight Coherence (PFC) reward. PFC leverages the V-JEPA2 world model as a physics referee to penalize implausible dynamics in the latent feature space. Experiments confirm MIND-V's SOTA performance in long-horizon simulation and its significant value for policy learning, introducing a scalable and fully autonomous framework for embodied data synthesis.