Generative Learning
of World (GLOW)

KnowinDream

Generative 3D World Data Engine

Generates higher-quality embodied data distributions equipped with rich 3D structure

Visual Input

Robot Cameras   (Images / Video)

Language

User Text Instruction

Robot State &
Execution Feedback

Proprioception / Rollout Traces

EMBODIED FUNDATION MODEL

KnowinBrain

(Post-trained for Action CoT)

Autoregressive CoT for aligned with action token generation

ACTION TOKENIZER

KnowinAct

(Pre-trained Action Decoder)

The Action token space is growing dynamically

RLHF

Action
Chunk

KnowinWorld

Interactive World Dynamics Sandbox

ACTION ROLLOUT

Robot Execution

Action chunks are converted into smooth arm motion locally

Embodied Alignment

World DynamicsWorld-grounded
rollout
Active ExplorationSkill composition
Interactive
Reasoning
Brain-act
consistency