SEMAG: Self-Evolutionary Multi-Agent Code Generation
This addresses the problem of inflexibility in automated code generation for developers, representing a strong incremental advance in multi-agent systems for programming tasks.
The paper tackles the limitation of manual model selection and fixed workflows in LLM-based code generation by proposing SEMAG, a self-evolutionary multi-agent framework that adapts to task complexity and automatically upgrades models, achieving a 3.3% improvement on CodeContests and up to 52.6% Pass@1 accuracy with optimal backbone selection.
Large Language Models (LLMs) have made significant progress in handling complex programming tasks. However, current methods rely on manual model selection and fixed workflows, which limit their ability to adapt to changing task complexities. To address this, we propose SEMAG, a Self-Evolutionary Multi-Agent code Generation framework that mimics human coding practices. It decomposes programming tasks into stages, including planning, coding, debugging, and discussion, while adapting workflows to task difficulty. Its self-evolutionary agents can access the latest models in real time and automatically upgrade the backbone model. SEMAG sets new state-of-the-art Pass@1 accuracy across benchmarks. Using identical backbone models, SEMAG outperforms prior methods by 3.3% on CodeContests. When augmented with self-evolutionary model selection that automatically identifies optimal backbones, SEMAG reaches 52.6%, showcasing both framework effectiveness and adaptability to evolving LLM capabilities.