HomeBody: Long-horizon humanoid autonomy without policy learning

Researchers at Caltech and Stanford present HomeBody, which lets a frontier VLM directly orchestrate composable motor skills, giving a Unitree G1 humanoid persistent spatial memory. In an unseen kitchen it cleans across the room and retrieves remembered objects from underspecified requests without environment-specific training data or policy learning.

• Study finds no significant AI displacement of new grads2026-09-25
• PrismML brings tiny LLMs to Qualcomm smart glasses2026-09-24
• Stanford and NVIDIA Release CLM-8B Contrastive Language Model2026-09-24