Research

Research Interests

Recommender systems and agentic AI for real-world applications.

My research focuses on recommender systems and agentic AI in various real-world applications.

  • Recommender Systems Multi-behavior recommendation, LLM-based Recommendation. LLM-based recommendation
    • Training-free Approach Recommending with LLMs without fine-tuning, refining their outputs with collaborative signals.
    • Reranking in LLM RecSys Reranking candidate items with LLMs for more accurate recommendation.
  • Agentic AI in Real-World Applications Building LLM agents that solve practical tasks across various domains. Representative applications
    • Relational Databases Relational database learning with LLM agents.
    • Ad Domain Data-driven ad ranking for practical advertising scenarios.

Publications

Selected Publications

Publications [C: Conference, J: Journal, T: Tutorial, P: Preprint]

2027

  1. [C16] User-Adaptive Fusion of Multiple Foundation Models for Recommender Systems

    Sunwoo Kim, Hyunjin Hwang, Geon Lee, Kyungho Kim, and Kijung Shin

    WSDM 2027 Foundation model for recsys

2026

  1. [C15] Zero-Shot Tabular Learning via Hypergraph Reasoning with LLMs

    Geon Lee, Kyungho Kim, Dongwon Choi, Juyeon Kim, Shinhwan Kang, and Kijung Shin

    EMNLP 2026 Tabular learning, LLM
  2. [C14] Training-Free LLM-Based Recommendation with Post-LLM Item Refinement Using Collaborative Signals

    Kyungho Kim, Sunwoo Kim, Geon Lee, Shinhwan Kang, Sojeong Kim, Liam Collins, Bhuvesh Kumar, Donald Loveland, and Kijung Shin

    CIKM 2026 (Short Paper) Training-free LLM Recsys, Refinement
  3. [P1] Retrieval-Augmented Generation for LLM-based Recommender Systems: A Comprehensive Survey

    Sunwoo Kim*, Geon Lee*, Kyungho Kim*, Liam Collins, Neil Shah, and Kijung Shin

    Preprint LLM Recsys, RAG
  4. [T1] A Tutorial on Retrieval-Augmented Generation for LLM-based Recommender Systems

    Sunwoo Kim*, Geon Lee*, Kyungho Kim*, Liam Collins, Neil Shah, and Kijung Shin

    KDD 2026, CIKM 2026, WSDM 2027 (Survey paper [P1]) LLM Recsys, RAG
  5. [C13] Rethinking Contrastive Learning for Graph Collaborative Filtering: Limitations and A Simple Remedy

    Geon Lee, Sunwoo Kim, Kyungho Kim, and Kijung Shin

    ICML 2026 Graph Collaborative Filtering, Contrastive Learning
  6. [C12] Pre-Deployment Advertisement Ranking under Data Scarcity via Context-Aware Criteria Generation with VLMs

    Kyungho Kim*, Yeonje Choi*, Gyurim Hwang, Sejin Chung, Hongseok Lee, Myeong Ho Song, Yeongho Kim, Sunwoo Kim, Jongha Lee, Juyeon Kim, and Kijung Shin

    ACL 2026 (Industry Paper) & DATA-FM workshop in ICLR 2026 VLM Agent, Marketing AI ACL Oral Presentation
  7. [C11] From Raw Features to Effective Embeddings: A Three-Stage Approach for Multimodal Recipe Recommendation

    Jeeho Shin, Kyungho Kim, and Kijung Shin

    SIGIR 2026 (Short Paper) Multimodal Recsys
  8. [C10] ItemRAG: Item-Based Retrieval-Augmented Generation for LLM-Based Recommendation

    Sunwoo Kim, Geon Lee, Kyungho Kim, Jaemin Yoo, and Kijung Shin

    SIGIR 2026 (Short Paper) LLM Recsys, RAG
  9. [C9] ReFuGe: Feature Generation for Prediction Tasks on Relational Databases with LLM Agents

    Kyungho Kim, Geon Lee, Juyeon Kim, Dongwon Choi, Shinhwan Kang, and Kijung Shin

    WWW 2026 (Short Paper) LLM Agent, Relational Database
  10. [J2] A Comprehensive Survey on Multi-Behavior Recommender Systems: Extended Taxonomy and Recent Advances

    Kyungho Kim, Sunwoo Kim, Geon Lee, Jinhong Jung, and Kijung Shin

    JDSA (International Journal of Data Science and Analytics) (Invited as one of the best-ranked papers of PAKDD 2025) Multi-behavior Recommendation
  11. [C8] Feature-Centric Unsupervised Node Representation Learning Without Homophily Assumption

    Sunwoo Kim, Sooyong Lee, Kyungho Kim, Hyunjin Hwang, Jaemin Yoo, and Kijung Shin

    AAAI 2026 Graph Representation Learning

2025

  1. [C7] RDB2G-Bench: A Comprehensive Benchmark for Automatic Graph Modeling of Relational Databases

    Dongwon Choi, Sunwoo Kim, Juyeon Kim, Kyungho Kim, Geon Lee, Shinwhan Kang, Myungwhan Kim, and Kijung Shin

    NeurIPS 2025 (Dataset & Benchmark) RDB Benchmark
  2. [C6] A Self-supervised Mixture of Experts for Multi-behavior Recommendation

    Kyungho Kim, Sunwoo Kim, Geon Lee, and Kijung Shin

    CIKM 2025 Multi-behavior Recommendation
  3. [J1] Revisiting LightGCN: Unexpected Inflexibility, Inconsistency, and A Remedy Towards Improved Recommendation (Extended Version)

    Geon Lee, Kyungho Kim, Fanchen Bu, Langzhang Liang, and Kijung Shin

    ACM TORS (Invitation) Graph Collaborative Filtering
  4. [C5] Multi-Behavior Recommender Systems: A Survey

    Kyungho Kim, Sunwoo Kim, Geon Lee, Jinhong Jung, and Kijung Shin

    PAKDD 2025 Multi-behavior Recommendation Best Survey Paper Award

2024

  1. [C4] Rethinking Reconstruction-based Graph-Level Anomaly Detection: Limitations and a Simple Remedy

    Sunwoo Kim, Sooyong Lee, Fanchen Bu, Shinwhan Kang, Kyungho Kim, Jaemin Yoo, and Kijung Shin

    NeurIPS 2024 Graph Anomaly Detection
  2. [C3] Revisiting LightGCN: Unexpected Inflexibility, Inconsistency, and A Remedy Towards Improved Recommendation

    Geon Lee, Kyungho Kim, and Kijung Shin

    RecSys 2024 (Short) Graph Collaborative Filtering Best Short Paper Candidate (Top 7)
  3. [C2] Post-Training Embedding Enhancement for Long-Tail Recommendation

    Geon Lee, Kyungho Kim, and Kijung Shin

    CIKM 2024 (Short) Long-tail Recommendation
  4. [C1] Towards Better Utilization of Multiple Views for Bundle Recommendation

    Kyungho Kim, Sunwoo Kim, Geon Lee, and Kijung Shin

    CIKM 2024 (Short) Bundle Recommendation

Background

Experience & Education

Research Experience

  • MS+Ph.D Student, Data Mining Lab
    KAIST AI · Advisor: Prof. Kijung Shin
    Recommender Systems, Large Language Model Applications, Relational Database Learning
    2024 — Present
  • Undergraduate Student Researcher
    KAIST AI · Data Mining Lab
    2023 Jun — 2024 Jul
  • Undergraduate Student Researcher
    Yonsei University · Data Intelligence Lab (Advisor: Prof. Dongha Lee)
    2023 Mar — 2023 May
  • Undergraduate Student Researcher
    Korea University · DSBA Lab (Advisor: Prof. Pilsung Kang)
    2022 Jun — 2022 Dec

Education

  • M.S. + Ph.D. in Artificial Intelligence
    KAIST, School of AI
    2024 Aug — Present
  • B.S. in Industrial & Management Engineering (Graduate with Great Honor)
    Korea University
    2020 Mar — 2024 Aug

Recognition

Awards & Honors

  • PAKDD 2025 Best Survey Paper Award2025
  • Graduated with Great Honor, Korea University2024
  • Academic Perfect Award (Dean’s Award), Korea University2023 Spring
  • Academic Excellence Award, Korea University2021, 2022
  • Full Scholarship for Academic Excellence, Korea University2021–2023
  • Korea University Leader Scholarship, Korea University2021

Service

Academic Service

  • ACM CIKM — Session Chair2025-2026
  • ACM WWW — Reviewer2025-2026
  • ACM TKDD — Reviewer2026
  • Information Processing and Management — Reviewer2026
  • ICLR — Reviewer2027

Leadership

Leadership & Activities

Student leadership and service roles.

  • President, Industrial & Management Engineering Student Council (Korea University)2020 Oct — 2021 Dec
  • Student Body President, Boin High School2017 Sep — 2018 Aug

Get in Touch

Contact

📧 Email: kkyungho@kaist.ac.kr
📍 Seoul, Republic of Korea