Rui Ge

Rui Ge

Undergraduate at Shanghai Jiao Tong University(SJTU)

About Me

I am Rui Ge, an undergraduate student at Shanghai Jiao Tong University (SJTU), where I am pursuing my studies with a focus on cutting-edge research in artificial intelligence and Large Language Models. My academic journey is driven by a passion for innovative problem-solving and advancing technology through collaborative and interdisciplinary approaches.

Research Experience

  • Adaptive Multi-Agent System Generation with Large Language Models (Nov 2024 - Feb 2025)
    As a Researcher under Prof. Siheng Chen at SJTU and Shanghai AI Lab, I proposed a novel framework for automating multi-agent workflow generation using large language models, enhancing structured coordination for complex reasoning tasks.

  • Towards Context-Aware LLM Agent: Data Pruning for Sensor Data Comprehension (Oct 2024 - Feb 2025)
    As the Main Researcher under Prof. Zhenzhe Zheng at SJTU, I designed a technique to map sensor data into semantic space with data pruning, reducing inference costs while enabling LLMs to interpret sensor data effectively.

  • Synthetic Data Generation from Unstructured Text for Federated Instruction Tuning (Jul 2024 - Oct 2024)
    As the Main Researcher under Prof. Siheng Chen, I developed FedIT-U2S, a framework transforming unstructured text into high-quality instruction data for federated learning, eliminating the need for manual annotation.

  • Real-World Benchmarks for Federated Large Language Models (Jan 2024 - Jun 2024)
    As the Main Researcher under Prof. Siheng Chen, I created FedLLLM-Bench, the first benchmark suite for federated large language models, integrating diverse corpora to evaluate performance under realistic constraints.

Honors and Awards

  • Rongchang Science and Technology Innovation Scholarship (¥10,000) - Shanghai Rongchang Public Welfare Foundation & SJTU, 2024
  • Zhiyuan Honor Program Scholarship (¥5,000 per year) - Zhiyuan College, SJTU, 2022, 2023, 2024
  • Reviewer of ICML 2025 Workshop, 2025

Skills

  • Languages: Mandarin Chinese (native), English (TOEFL 108/120)
  • Programming: Python, C++, C
  • Technologies: Instruction Fine-Tuning, Direct Preference Optimization (DPO), Prompt Engineering, Multi-Agent System Design, Web Scraping
  • Tools: PyTorch, Transformers, LaTeX, Linux

I am eager to collaborate on projects that push the boundaries of AI research and application. Feel free to reach out via email at motigrez@sjtu.edu.cn for any opportunities or discussions!