When RAG Meets LFMs: Towards Retrieval-Augmented Large Foundation Models
Tutorial for PAKDD 2026
Tell Me More!

About

As one of the most advanced techniques in AI, Retrieval-Augmented Generation (RAG) can provide reliable and up-to-date external knowledge, offering significant convenience for numerous tasks. Particularly in the era of AI-generated content (AIGC), RAG's powerful retrieval capacity provides additional knowledge, enabling retrieval-augmented generation to assist existing generative AI in producing high-quality outputs. Recently, Large Foundation Models (LFMs) have demonstrated revolutionary abilities in language understanding and generation, while still facing inherent limitations such as hallucinations and out-of-date internal knowledge. Given RAG's ability to provide the latest and most helpful auxiliary information, retrieval-augmented large foundation models have emerged to harness external, authoritative knowledge bases rather than solely relying on a model's internal knowledge to improve generation quality.

In this tutorial, we comprehensively review existing research studies in retrieval-augmented large foundation models (RA-LFMs), covering three primary technical perspectives: architectures, training strategies, and applications. As preliminary knowledge, we briefly introduce the foundations and recent advances of LFMs. Then, to illustrate the practical significance of RAG for LFMs, we categorize mainstream relevant work by application area, detailing the challenges of each and the corresponding capabilities of RA-LFMs. Finally, to deliver deeper insights, we discuss current limitations and several promising directions for future research.

Related prior tutorials: KDD 2024 and ICDE 2025

Slides: PAKDD 2026 Tutorial Slides


TARGET AUDIENCE AND PREREQUISITES FOR THE TUTORIAL

This tutorial targets college students, academic researchers, and industrial practitioners interested in Retrieval-Augmented Generation (RAG) and Large Foundation Models (LFMs). Participants are expected to have basic knowledge of artificial intelligence, language models, and retrieval techniques. The tutorial is designed to be accessible to junior/senior undergraduate level audiences while remaining valuable to experienced researchers and engineers who want a structured entry point. After attending, participants are expected to gain a comprehensive understanding of RA-LFMs and learn how to design solutions for customized problems.

Event Dates

PAKDD 2026 Tutorial Session (Date and time: June 9, 2026, 13:30-17:00)

Tutorial Syllabus

The topics of this tutorial include (but are not limited to) the following:

  • Retrieval-Augmented Generation (RAG)
  • Large Foundation Model (LFM)
  • Information Retrieval
  • Pre-training & Fine-tuning
  • Agentic AI
  • Trustworthy RAG

    The tutorial outline is shown below:

  • Part I: Introduction (10 minutes)
  • Part II: Architecture of Combining RAG with LFMs (70 minutes)
    • RA-LFM architecture overview
    • Retriever in RA-LFMs
    • Retrieval results integration
    • Pre/Post-retrieval techniques
    • Special RA-LFM paradigms
  • Coffee Break (30 minutes)
  • Part III: Learning of RAG (30 minutes)
    • Training-free methods
    • Training-based methods
      • Independent Learning
      • Sequential Learning
      • Joint Learning
  • Part IV: Agentic RAG (20 minutes)
    • Prompting-based RAG
    • Single-Agent RAG
    • Multi-Agent RAG
    • RL-Driven RAG
  • Part V: Applications of RAG (25 minutes)
    • Preliminary applications in NLP
    • Downstream tasks
    • Domain-specific applications (e.g., science, finance, law)
  • Part VI: Future Directions and Challenges (10 minutes)
    • Trustworthy RA-LFMs
    • Multi-Modal RA-LFMs
    • Efficient and Scalable RA-LFMs
    • Autonomous RA-LFMs
    • Domain-Adaptive and Personalized RA-LFMs
  • Q&A (15 minutes)

    • Organization


    Tutorial TUTORS

    Xu Yuan

    PhD Student

    The Hong Kong Polytechnic University

    Yujuan Ding

    Research Assistant Professor

    The Hong Kong Polytechnic University

    Chengliang Liu

    PhD Student

    The Hong Kong Polytechnic University

    Rui An

    PhD Student

    The Hong Kong Polytechnic University

    Chun-Hin Chan

    MPhil Student

    The Hong Kong Polytechnic University

    Yiqi Wang

    Assistant Professor

    National University of Defense Technology

    Wenqi Fan

    Assistant Professor

    The Hong Kong Polytechnic University

    Qing Li

    Chair Professor and Head

    The Hong Kong Polytechnic University