A world model predicts how the world will change. In robotics, it learns the consequence of a proposed action rather than directly deciding which action the robot must execute.

Policy compared with world model

ModelMain mappingDeployment behaviour
VLA policyobservation + instruction → actionSends an action or action chunk to the robot
World modelcurrent latent state + candidate action → future latent stateLets a planner compare possible consequences before execution

RT-1, RT-2, OpenVLA, and SmolVLA are policies. They learn to output actions from observations and instructions. A world model instead learns cause and effect:

current state + candidate action → predicted next state

Planning loop

  1. Propose several candidate action sequences
  2. Use the world model to perform a Latent Rollout for each sequence
  3. Score the predicted result against the goal
  4. Select the best sequence
  5. Execute the first action
  6. Observe the new state and plan again

The Cross-Entropy Method is one way to search the candidate sequences.

Generative and predictive world models

  • A generative world model predicts pixels or visual tokens that can be decoded into pixels
  • A predictive world model predicts latent representations and avoids pixel generation

V-JEPA 2 and GeoWorld are predictive world models. Their planner compares predicted future representations with a goal representation.

Important distinction

A world model is a role, while Joint-Embedding Predictive Architecture (JEPA) is an architecture family. A JEPA can support a world model, but the two terms do not mean the same thing.

My way of understanding it

The clearest comparison for me is with SmolVLA. SmolVLA is a policy that directly maps its observations and instructions to an action chunk, which is then output to the physical robot. However, V-JEPA 2-AC is a world model that takes a candidate action sequence and predicts the future latent representation that would result from those actions.

In a way, we are trying to let the model understand the physics of our world and learn cause and effect through experience. Instead of immediately asking, "What action should I execute?", the world model asks, "If I take this action, what will happen next?"

This is why I think of a policy as the model that acts and a world model as the model that imagines the consequence before acting.