Two models are compared: Model A is a high-degree polynomial with no regularization; Model B is a low-degree polynomial with no regularization. A third model C applies strong regularization to Model A's architecture. How does Model C's bias-variance profile likely compare to Model B?