<p>Multimodal multiobjective optimization problems (MMOPs) frequently arise in real-world applications, where widely separated solutions in the decision space may yield similar objective values. The principal challenge in solving MMOPs lies in identifying all equivalent Pareto-optimal solutions while maintaining an appropriate balance between diversity and convergence in both decision and objective spaces. This paper proposes a novel algorithm, Dynamic Topology Multiobjective Particle Swarm Optimization with Adaptive Lévy Flight (DT-MMPSO-Levy), to address these challenges. DT-MMPSO-Levy incorporates a dynamic small-world topology, enabling adaptive re-wiring of particles throughout generations. This facilitates both local and global information exchange, enhancing decision space diversity. To balance exploration and exploitation across the optimization process, a phase-based adaptive velocity update mechanism is employed. During early generations, Lévy flight-induced heavy-tailed jumps promote global exploration for discovering diverse Pareto-optimal sets. In mid generations, an adaptive control parameter, Omega, modulates the exploration-exploitation trade-off based on population diversity. In later stages, the algorithm shifts focus toward convergence by emphasizing local exploitation. The performance of DT-MMPSO-Levy is benchmarked against seven state-of-the-art MMOP algorithms on twelve standard test problems from the CEC 2019 suite. Experimental results show that DT-MMPSO-Levy achieves superior performance, securing the best scores in 8 out of 12 cases for Inverted Generation Distance in Decision Space (IGDX), 7 out of 12 for Pareto Sets Proximity (PSP), and 7 out of 12 for Inverted Generation Distance in Objective Space (IGDF) metrics, thereby demonstrating its effectiveness and robustness.</p>