Multi-agent large language model frameworks are promising for complex multi step reasoning, yet existing systems remain weak for scientific and knowledge intensive domains due to static prompts and agent roles, rigid workflows, and homogeneous model reliance, leading to poor domain adaptation, limited reasoning flexibility, and high latency on heterogeneous or long-horizon scientific tasks. They also struggle to revise earlier decisions when intermediate reasoning diverges, reducing reliability in structured and calculation heavy settings. To address these limitations, we propose a scientific domain oriented interactive two tier multi model orchestration framework. A dedicated orchestration model analyzes each task, dynamically constructs a domain aware reasoning pipeline, and instantiates specialized expert agents with tailored prompts, while an execution model performs each step under generated role and instruction specifications. The orchestrator iteratively updates the pipeline based on intermediate feedback, enabling dynamic replanning, role reallocation, and prompt refinement across multi turn interactions, strengthening robustness and specialization for scientific reasoning through structured heterogeneous model collaboration. The framework is model agnostic and supports heterogeneous LLM integration with different capacities or costs, enabling flexible performance efficiency trade offs in practical scientific deployments. Experiments show consistent improvements over existing multi agent systems and strong baselines across diverse reasoning and scientific style benchmarks.
Many Heads Are Better Than One: Improved Scientific Idea Generation by A LLM-Based Multi-Agent System
Shanghai Artificial Intelligence Laboratory（上海人工智能实验室）；Shanghai Innovation Institute（上海创新研究院）；Department of Engineering Science, University of Oxford（牛津大学工程科学系）；Shanghai Institute for Science of Science（上海科学研究院）；Department of Information Engineering, Chinese University of Hong Kong（香港中文大学信息工程系）；Department of Electronic Engineering, Tsinghua University（清华大学电子工程系）
CXRAgent: Director-Orchestrated Multi-Stage Reasoning for Chest X-Ray Interpretation
School of Computer Science, Hangzhou Dianzi University（杭州电子科技大学计算机科学学院）；Department of Automation, Tsinghua University（清华大学自动化系）；Department of Colorectal Medical Oncology, Zhejiang Cancer Hospital（浙江省肿瘤医院结直肠医学肿瘤科）；School of Computer and Control Engineering, University of Chinese Academy of Sciences（中国科学院大学计算机与控制工程学院）；School of Intelligence Science and Engineering, Harbin Institute of Technology (Shenzhen)（哈尔滨工业大学（深圳）智能科学与工程学院）
Collaborative Agent Reasoning Engineering (CARE): A Three-Party Design Methodology for Systematically Engineering AI Agents with Subject Matter Experts, Developers, and Helper Agents
