Qi Zhu

Table of Contents

Quick Links

Applied Scientist
Think Forward Lab
AWS AI/ML Services & Infrastructure
Email: qi.zhu.ckc@gmail.com

Bio
I obtained my Ph.D. in Computer Science from University of Illinois at Urbana-Champaign advised by Prof. Jiawei Han, where I was a member of Data and Information Systems Laboratory (DAIS) and Data Mining Group. Here is my CV (slightly outdated).

Research

At AWS, I design AI systems (GraphStorm) utilizing structured knowledge for applications in retrieval-augmented generation (RAG), graph machine learning, and beyond. As a founding member of the GraphRAG Team, I help drive the launch of structure-aware features across various RAG services.

My current and past work focuses on the following themes:

  1. LLMs with Structured Knowledge – Harnessing explicit and implicit data structures to enhance the long-context performance and parameter efficiency of large language models.
  2. Graph Representation Learning – Representing objects in heterogenous text-attributed graph with heterogenous learning, and robust to distribution shift.

I. LLMs with Structured Knowledge

We aim to make LLMs more efficient and resilient against hallucinations by harnessing structured knowledge. A key challenge lies in making the language model structure-aware while mitigating performance bottlenecks, such as the lost-in-the-middle phenomenon, To address this, we explore post-training, fine-tuning, and pre-training techniques on graph structured data.

  • Graph Retrieval Augmented Generation: We develop structure-aware algorithms for pre-retrieval, retrieval, and inference stages of RAG.
    • AGENT-G: An Agentic Framework for Graph Retrieval Augmented Generation, Under Review

II. Graph Representation Learning

My research aims to make graph representation learning adapt to distribution shift and data heterogeneity.

Awards

  • 2020 Amazon AWS Machine Learning Research Award
  • 2018 ACM WWW Best Poster Honorable Mention