What are preferences and preference functions in multi-objective reinforcement learning?

In RL (reinforcement learning) or MARL (multi-agent reinforcement learning), we have the usual tuple:

(state, action, transition_probabilities, reward, next_state)


In MORL (multi-objective reinforcement learning), we have two more additions to the tuple, namely, "preferences" and "preference functions".

What are they? What do we do with them? Can someone provide an intuitive example?