In this presentation, I will initially discuss my published work titled “A constrained multi-objective evolutionary algorithm with clustering based weight vector adaptation” The talk will highlight the methodologies developed to efficiently handle constrained multi-objective optimization problems.
Subsequently, I will present my current research efforts focused on employing evolutionary algorithms for optimizing model merging configurations of Large Language Models (LLMs). Lastly, I will outline my future research directions, particularly focusing on the potential of leveraging LLMs to design and enhance evolutionary algorithms, and exploring new frontiers in optimization strategies and methodologies.
