Jul 2026· 2026 6th International Conference on Intelligent Communications and Computing (ICICC)· pp. 399-402· 0 citations· 14 references
Abstract
While Transformer-based Large Language Models (LLMs) have advanced NLP, achieving an efficient automated development workflow in multi-agent systems remains challenging due to issues in code accuracy, testing effectiveness, and agent collaboration. To address these limitations, we propose UMLCoder, a novel multi-agent collaborative code generation framework designed to enhance software reliability and maintainability. UMLCoder comprises four specialized agents: a UML Expert Agent for generating precise structural diagrams to guide programming, a Code Generation Agent, a Test Case Generation Agent, and a Test Execution Agent for robustness verification. Benchmark evaluations demonstrate that UMLCoder achieves a pass@1 score of 71.9% on HumanEval and 70.3% on MBPP. Compared to baselines like GPT-3.5-turbo and LLaMA3, the proposed framework significantly improves code quality and computational efficiency, reducing the time complexity from $O\left(n^{2}\right)$ to $O(n)$ in specific scenarios.
TestAgent, a multi-agent tool implemented as a VS Code extension that automates the generation of high-quality unit tests for Java projects using repository-level Code Knowledge Graphs, demonstrates its practical utility for regression testing and bug discovery.
Ye Shang, Quanjun Zhang, Zheng Zhan et al.· SIGSOFT FSE Companion· 0 citations
This work proposes TraceDev, a multi-agent framework for automated software development grounded in use cases that contain multiple functional points and complex semantics, and demonstrates the effectiveness of TraceDev in repository-level code generation from requirements.
Mingyu Chen, Yakun Zhang, Zihao Xie et al.· 0 citations
Contemporary large language models can generate syntactically correct code from natural language specifications. However, integrating the generated code into existing software projects remains a largely manual, error-prone process that severely limits practical utility. This paper proposes a novel four-agent multi-agent system architecture to bridge the gap between code generation and code integration. The system comprises a Context Retrieval Agent, Code Generation Agent, Code Integration Agent, and Orchestration Agent. Evaluated through 270 experiments across three software projects of varying complexity, six language models, and three code generation styles, the AST-based integration architecture achieved a 100% success rate. Syntax validity averaged 69.2%, and the best model configuration reached a normalized score of 0.795.
Yasin Büzgülü, Selim Akyokuş, S. Sarp· Signal Processing and Commun...· 0 citations
The AgentCodeReview system is presented, a multi-agent system that is able to conduct explainable code review and automated bug repair by leveraging software engineering agents with different code review tasks and its utility and extensibility to the field of explainable AI in software quality assurance are demonstrated.
B. N, T. L. Manasa· International journal of com...· 0 citations
Cross-platform code generation is a critical challenge in modern software engineering, particularly for systems targeting heterogeneous instruction set architectures and operating systems. Although large language models (LLMs) have demonstrated strong capabilities in code generation, existing approaches often suffer from implicit semantic modeling, insufficient handling of platform-specific constraints, and lack of reliable validation mechanisms, leading to unstable and non-compilable outputs in real-world scenarios. To address these limitations, this paper proposes CCGMAS, a multi-agent framework for cross-platform Go code generation. The framework consists of four collaborative agents: a Platform Residue Analysis Agent that extracts platform-dependent semantics and constructs a structured platform residue profile, a Requirement Generation Agent that transforms multi-source code context into a requirement-centered semantic representation with iterative quality optimization, a Code Generation Agent that produces target-platform implementations through multi-candidate synthesis, and a Verification Agent that performs progressive validation and provides structured feedback. By introducing requirement documents as an intermediate semantic layer and incorporating platform residue modeling, CCGMAS enables more explicit semantic alignment across platforms. Furthermore, a feedback-driven refinement loop is designed to iteratively correct errors at different stages, improving both functional correctness and platform compatibility. Experiments on CCGBench demonstrate that CCGMAS reduces the platform residue rate by up to 62.2% in the best case and improves the cross-platform build pass rate by up to 45.0% compared with direct translation baselines.
Large Language Models (LLMs) offer new opportunities for automated code refactoring. However, generated changes must reduce targeted quality problems without introducing new issues or altering behaviour-relevant code structures. We introduce REFINE (Refactoring with Evidence-aware Flow for Integrated ageNtic Execution), a tool-agnostic, evidence-aware multi-agent approach for generating Java file-level refactoring candidates. REFINE combines static-analysis-guided smell identification, smell-informed planning, LLM-based transformation, automated re-analysis, preservation checks, and structured reporting. We evaluate REFINE on 450 Java files from 15 open-source systems, producing 1,350 model-pass outputs using OpenAI GPT-5.5, Google Gemini 3.1 Pro Preview, and Anthropic Claude Opus 4.8. REFINE reduces detected code smells by 68.26%, 72.79%, and 68.49% across the three configurations, respectively, with the strongest reductions observed for major smells. A matched 150-file direct-prompt baseline shows that REFINE achieves a higher median code-smell reduction with smaller edits and fewer public-method removals. However, broader quality improvements are inconsistent, and preservation checks reveal residual risks, including assert/fail-call changes and public-method removal. Therefore, REFINE outputs should be treated as refactoring candidates requiring compilation, testing, dependency analysis, and human review before adoption in repository- or system-level settings.
Muhammad Waseem, Aakash Ahmad, Pekka Abrahamsson· 0 citations