What does the action$-$value function $\backslash$pi $(,)$ represent in Reinforcement Learning?

Group of answer choices

The expected reward received after taking action $$ in state $$ and following a specific policy thereafter.

The immediate reward received after taking action $$ in state $$

The probability of receiving a reward after taking action $$ in state $$

The expected cumulative reward starting from state $$ and following a particular policy.