Results for "collective behavior"

AI tacitly coordinating prices.

Internal representation of the agent itself.

Inferring and aligning with human preferences.

Updating a pretrained model’s weights on task-specific data to improve performance or adapt style/behavior.

Learning only from current policy’s data.

System-level design for general intelligence.

Training a smaller “student” model to mimic a larger “teacher,” often improving efficiency while retaining performance.