Get Started
Home
Topics
Search
Library
Research questionHow can language-driven humanoid control follow instructions while staying physically plausible and stable over long horizons?Natural-language commands must be converted into coordinated humanoid actions that respect physical dynamics. Small control errors can accumulate over long rollouts, causing unstable behavior or motion that no longer matches the instruction.
AI
Diffusion Models
Reinforcement Learning
Research Paper
Robotics
Latest papersRecent research connected to this question, newest first.SCRIPT: Scalable Diffusion Policy with Multi-stage Training for Language-driven Physics-Based Humanoid ControlThe source studies physics-based humanoid control using action, physical-state, and text inputs, with diffusion-policy training and reinforcement learning in closed-loop simulations. Evidence includes quantitative comparisons across instruction alignment, motion quality, and physical realism, plus scaling studies on the 1,200-hour MotionMillion dataset; real-world deployment is not covered.research paper · Sep 4, 2026
Related questions
How can humanoid morphology and whole-body control be co-designed to reproduce human movement?How can vision-language models ground semantic driving inputs in physically plausible continuous actions with low latency?How can language-conditioned robots preserve executable actions under irrelevant wording while composing multiple changed constraints correctly?How can vision-language-action systems reliably execute long-horizon manipulation while tracking state and conditional dependencies?