The field of optimization algorithms and meta-learning is rapidly advancing, with a focus on developing more efficient and effective methods for solving complex problems. Recent research has explored the use of novel metaheuristic approaches, such as single-point optimization and resource-centered modeling, to tackle discrete and continuous optimization problems. Furthermore, the development of unified benchmark platforms and automated algorithm design frameworks has enabled the systematic evaluation and improvement of optimization algorithms. Notable advancements include the creation of frameworks that can generate effective optimizers for specific problems, as well as the introduction of new algorithms that outperform existing methods in terms of efficiency and adaptability.

Noteworthy papers include: DesignX, which presents an automated algorithm design framework that generates effective optimizers for black-box optimization problems. MetaBox-v2, which introduces a unified benchmark platform for meta-black-box optimization that supports various approaches and provides a comprehensive suite of tasks. LOKI, which proposes a compute-efficient framework for co-designing morphologies and control policies that generalize across unseen tasks. SEvoBench, which provides a modern C++ framework for evolutionary computation and benchmarks evolutionary single-objective optimization algorithms.