EvoX：用于自动发现的元进化

EvoX introduces a novel approach that combines LLM-driven optimization with evolutionary search to effectively enhance programs, prompts, and algorithms across various domains. In this paradigm, previously evaluated solutions are reused to guide the model toward new candidate solutions. The effectiveness of this evolutionary process hinges on the search strategy, specifically how prior solutions are selected and varied to generate new candidates. However, most existing methods rely on fixed search strategies with predefined parameters (e.g., explore-exploit ratios) that remain static throughout execution. While effective in certain contexts, these approaches often struggle to adapt across different tasks or even within the same task as the search space evolves. EvoX aims to address this issue by dynamically adjusting the search strategy, thereby improving the flexibility and efficiency of evolutionary search.