KCoEvo: A Knowledge Graph Augmented Framework for Evolutionary Code Generation

2026-03-08

AI Summary

This work addresses the challenge of code breakage caused by frequent third-party API changes, a problem exacerbated by large language models’ (LLMs) limited structural understanding of API evolution. To overcome this, the authors propose a knowledge graph–enhanced framework that jointly constructs static and dynamic API knowledge graphs to model both intra-version structural relationships and inter-version evolutionary transitions. Code migration is formulated as two synergistic stages: evolutionary path retrieval and path-guided code generation. The framework leverages synthetically generated supervision signals derived from real-world API differences to enable end-to-end training. Experimental results demonstrate that the approach significantly outperforms standard LLMs on both single-package and multi-package benchmarks, achieving notable improvements in migration accuracy, controllability, and execution success rate.

Technology Category

• Natural Language Processing: Code Generation / Program Synthesis from Natural Language
• Machine Learning: Large Multimodal Models (LMMs)
• Planning, Routing, and Scheduling: Planning with Language Models

Application Category

• Graph Algorithms and Modeling for the Web: Foundation models and LLMs for Web-related graphs
• Semantics and Knowledge: Methods to enhance, augment, integrate or synergize semantic models such as knowledge graphs and LLMs
• User Modeling, Personalization and Recommendation: Large Language Models (LLM) for user modeling and recommendation

Abstract

Code evolution is inevitable in modern software development. Changes to third-party APIs frequently break existing code and complicate maintenance, posing practical challenges for developers. While large language models (LLMs) have shown promise in code generation, they struggle to reason without a structured representation of these evolving relationships, often leading them to produce outdated APIs or invalid outputs. In this work, we propose a knowledge graph-augmented framework that decomposes the migration task into two synergistic stages: evolution path retrieval and path-informed code generation. Our approach constructs static and dynamic API graphs to model intra-version structures and cross-version transitions, enabling structured reasoning over API evolution. Both modules are trained with synthetic supervision automatically derived from real-world API diffs, ensuring scalability and minimal human effort. Extensive experiments across single-package and multi-package benchmarks demonstrate that our framework significantly improves migration accuracy, controllability, and execution success over standard LLM baselines. The source code and datasets are available at: https://github.com/kangjz1203/KCoEvo.

Problem

• code evolution
• API migration
• knowledge graph
• large language models
• software maintenance

Innovation

• knowledge graph
• code evolution
• API migration
• structured reasoning
• synthetic supervision