Hello! My name is Zhengliang Shi. I am now a second-year Master student majoring in Computer Science and Technology from General Artificial Intelligence Lab of Shandong University. During my master period, I fortunate to be advised by Zhaochun Ren and Pengjie Ren. Previously, I was supervised by Dr. Shuo Zhang in Bloomberg. Besides, I am fortunate to closely work with Dr. Weiwei Sun in Carnegie Mellon University. I obtained my Bachelor's degree in Shandong University majoring in Computer Science and Technology (graduated in 2023) and my research advisor is Zhaochun Ren.

📖 I work broadly on natural language processing and information retrieval, with a particular emphasis on large language models (LLMs). I investigate the core limitations of LLMs—such as hallucinations—that cannot be overcome by scaling alone. To address these challenges, my previous studies pioneered Retrieval-Augmented LLMs (and Tool-augmented LLMs), a novel class of LLMs that can automatically use diverse external tools during inference. Below, I briefly list the main research I have done before.

• Retrieval-augmented Generation. Enhancing models with factual knowledge from diverse sources of information and reduce their hallucination.
• Tool-learning Agent. Building real-world agents based on the foundation model to solve complex and practical tasks, e.g., tool-use agents.