MIND: Multi-Scale Intent Diffusion for Text-Driven Physics-Based Humanoid Control

2026-05-25

AI Summary

This work addresses the challenges in generating physically plausible humanoid motions from text, which are hindered by domain shifts between motion generation and physical control, as well as the modality gap between textual descriptions and motion dynamics. To bridge this gap, the authors propose an end-to-end diffusion framework that leverages hierarchical behavioral intent as a semantic intermediary. Specifically, a global intent predictor captures overall behavioral dynamics, while an instantaneous intent predictor refines local control, with structured inductive biases injected throughout the diffusion process. Humanoid states are encoded into a latent space and integrated with multi-scale intent prediction modules to enable direct generation of physically grounded motions from textual instructions. Experiments demonstrate that the proposed method significantly outperforms existing approaches in semantic alignment, motion naturalness, and physical plausibility.

Abstract

Enabling physics-based humanoids to execute diverse behaviors from high-level textual commands remains a significant challenge. Existing methods typically follow either a two-stage paradigm that combines kinematic motion generation with physics-based tracking, or an end-to-end imitation-learning paradigm that directly generates actions from text. However, the former suffers from the inherent domain shift between kinematic generation and physics-based tracking, while the latter struggles with the substantial modality gap between textual commands and low-level actions, limiting effective semantic alignment. Notably, humanoid states encode rich motion dynamics that are more semantically aligned with textual descriptions than low-level actions, making them a natural basis for deriving behavioral intent. Building upon this insight, we propose MIND, a novel end-to-end diffusion framework for text-driven physics-based humanoid control that leverages behavioral intent as a semantic bridge between textual commands and low-level actions. At its core, MIND introduces a multi-scale intent diffusion mechanism, where a holistic intent predictor captures global behavioral dynamics to guide overall behavior synthesis, while an immediate intent predictor provides step-wise, fine-grained signals for local behavior refinement at each diffusion step. This hierarchical intent formulation imposes a structured inductive bias for humanoid control, improving semantic alignment and behavioral naturalness. Furthermore, MIND encodes humanoid states into a latent space to enable more effective semantic intent modeling. Extensive experiments demonstrate that MIND outperforms existing methods and synthesizes coherent, physically plausible, and semantically aligned humanoid behaviors from text commands. Our code will be released to facilitate future research.

Problem

• text-driven control
• physics-based humanoid
• semantic alignment
• behavioral intent
• modality gap

Innovation

• multi-scale intent diffusion
• text-driven humanoid control
• behavioral intent
• semantic alignment
• physics-based animation