The field of reinforcement learning is moving towards more dynamic and complex tasks, with a focus on exploration and adaptation. Researchers are developing new methods to improve the efficiency and effectiveness of reinforcement learning in areas such as robotics and medical applications. One of the key challenges in these areas is the need for accurate and efficient exploration, which is being addressed through the development of new algorithms and frameworks. The use of uncertainty-driven adaptive exploration and task-informed rewards is showing promising results in improving the performance of reinforcement learning agents. Additionally, the application of reinforcement learning in medical areas such as cryoablation planning is demonstrating potential for improving treatment outcomes and reducing variability. Notable papers in this area include: Poke and Strike: Learning Task-Informed Exploration Policies, which proposes a task-informed exploration approach that achieves a 90% success rate on a striking task. Cryo-RL: automating prostate cancer cryoablation planning with reinforcement learning, which introduces a reinforcement learning framework that models cryoablation planning as a Markov decision process and learns an optimal policy for cryoprobe placement, achieving over 8 percentage-point Dice improvements compared with automated baselines.