The field of planning and exploration is moving towards more efficient and effective methods for trajectory exploration and optimization. Researchers are focusing on developing frameworks that can reason over complete plan compositions, rather than individual trajectories, and are exploring the use of diffusion models and uncertainty estimation to improve exploration and learning efficiency. Notable papers include Compositional Monte Carlo Tree Diffusion, which introduces a framework for globally-aware planning, and DreamerV3-XP, which improves exploration and learning efficiency through prioritized replay buffers and intrinsic rewards. Mixed Density Diffuser and Deep Active Inference with Diffusion Policy and Multiple Timescale World Model are also noteworthy for their innovative approaches to planning and exploration. Off-policy Reinforcement Learning with Model-based Exploration Augmentation is another significant contribution, which addresses the limitation in passive exploration through the generation of under-explored critical states and synthesis of dynamics-consistent experiences.