MeLA: A metacognitive LLM-driven architecture for automatic heuristic design

This paper introduces MeLA, a Metacognitive LLM-Driven Architecture that presents a new paradigm for Automatic Heuristic Design (AHD). Traditional nature evolutionary methods operate directly on heuristic code. Existing prompt-evolution methods mainly optimize task descriptions before generation. In contrast, MeLA evolves the instructional prompts used during heuristic generation and focuses on reflective guidance from previous outputs. This new paradigm, termed Metacognitive Prompt Evolution , is driven by a novel metacognitive framework where the system analyzes performance feedback to systematically refine its generative strategy. MeLA’s architecture integrates a problem analyzer to construct an initial strategic prompt, an error diagnosis system to correct faulty code, and a metacognitive search engine that iteratively optimizes the prompt based on heuristic effectiveness. In comprehensive experiments across both benchmark and real-world problems, MeLA achieves competitive performance on classical tasks and demonstrates clear advantages on more complex real-world problems. Ultimately, this research demonstrates the profound potential of using cognitive science as a blueprint for AI architecture, revealing that by enabling an LLM to metacognitively regulate its problem-solving process, we unlock a more robust and interpretable path to AHD.