Core Research Members
The DY-LLM team is composed of a group of interdisciplinary experts dedicated to building and optimizing large-scale generative AI models. The team includes members in the following fields: natural language processing experts, machine learning and deep learning engineers, data scientists, process experts, algorithm researchers, etc. The main goal is to develop models that can process large-scale natural language data and apply these models to practical scenarios, such as text generation, image generation, etc.
Chuanyou,Li(team)
Southeast University PhD Professor
Chief Scientist
Field of study：Artificial Intelligence and Machine Learning
Main contributions：He has published many papers in top journals and led the research on the underlying algorithm system of AI large models. Dr. Li focuses on AIGC, high-performance computing for AI model acceleration, combinatorial optimization problems (such as resource allocation, job scheduling, etc.), and system architecture in large language models (LLM). He has obtained the National Natural Science Foundation and the National Key Laboratory of Mathematical Engineering and Advanced Computing.
Kehua,Chan
University of Electronic Science and Technology of China PhD
AI Agent Development Engineer
Field of study：Software Engineering Government Information
Main contributions：Participate in national key scientific research projects, be responsible for the research and development of key technologies, focus on the innovative application of artificial intelligence technology, especially in the implementation and optimization of government systems. Promote the application of AI technology in government systems.
唐国雄
美国宾夕法尼亚大学 博士
深度学习工程师
研究领域：机器学习
主要贡献：主导开发了一系列高效的机器学习模型和算法，致力于提升模型的计算效率、准确性和可扩展性。这些创新的算法不仅在复杂数据处理和深度学习领域取得了显著成果，还有效解决了实际应用中的多项技术难题。
张文馨
法国雷恩大学 博士
中国科学院研究员
LLM产品总监
研究领域：现代管理理论与方法
主要贡献：现代企业管理的流式工程设计与管理、中国科学院院地战略研究。流式工程设计与管理成为推动地方经济高质量发展的重要动力。在全球化和数字经济的背景下，这一研究方向有望在未来为中国企业的持续发展提供更多创新的管理思路和实践指导。
梁苧芋
英国格拉斯哥大学 硕士
数据分析工程师
研究领域：数据挖掘
主要贡献：主导开发了一套高效的数据读取、清洗、分析和可视化系统，为企业和科研机构的数据处理工作提供了全方位的技术支持。该系统具备强大的数据处理能力，能够快速读取和解析大规模、异构化的数据源，大幅减少了数据处理的时间成本。
邢斯达
澳大利亚迪肯大学 硕士
LLM算法工程师
研究领域：循环神经网络(RNN)
主要贡献：针对传统循环神经网络(RNN)中梯度消失和梯度爆炸问题，开发一种新型LSTM算法——极限长短期记忆网络(E-LSTM)。该方法通过整合极限学习机(ELM)的逆矩阵部分为LSTM结构新增“门”机制，旨在减少训练轮次同时提高数据准确性，以提升文本预测等多个领域的效率。