INTERACTIVE ARCHITECTURE / FOURROOM

What goes into each component?

Open the spatial model explorer ↗ · Search the component index ↓

Pick a path, then a component. Brackets show tensor shape, not measured values. Shapes omit singleton batch axes.

INFERENCE / SEE THE SCENE · STEP 1

One picture becomes three sheets of numbers

MODEL ATLAS / SEARCH & NAVIGATE

Component map

Follow either input stream: pixels describe the scene; recorded commands describe movement. The planner searches trial codes rather than copying the training example.

Whole component map · stages in path orderSolid = animation · dashed = I/O diagram

Each row reads input → component → output. Select any box to locate it. “I/O only” opens its contract diagram; “Animated” opens the illustration.

NARRATED ANIMATION · KOKORO / GEORGE

From pixels to a turn

Goal

Read each panel top to bottom: input → operation → output. Shapes omit single batch axes.

START HERE / ONE TRIP THROUGH FOUR ROOMS

Can the dot reach the ring?

One try is called an episode. It starts when the simulator places the agent and sets a goal. A move is just one action inside that try.

GoalStart

SLOW MOTION / FIRST DECISION

Zoom in: why does move 21 differ in two runs?
FOURROOM / MODEL ANATOMY

Where does a turn come from?

EPISODE TIMELINE · WHY ZOOM IN HERE?
① Moves 1–10
Normal planner
② Moves 11–20
Same first target
③ Frame 20
Same scene · next target differs
④ Move 21 ↙
First different x/y request

Who produces move 21?

“21” is the move’s position in the recording—not a model input. Follow the arrows to see how the lower planner produces it.

A · SAVED TARGET
B · REPLANNED

What was scored?At step 10, plan 792 scored 315.64: squared difference between its predicted final features and the goal features. The researchers picked it for this comparison, not because it won. That score belongs to the full saved plan, not to its first target plus later replanning.
Actual observation images, requested commands and positions · g13-p43 · diagrams schematic · ◎ goal