Lianghao Xia (夏良昊)

Research Assistant Professor, University of Hong Kong

Email GitHub Google Scholar Personal CV
Foundation Models    Graph Learning    Recommendation Systems
Large Language Models    Self-Supervised Learning    Spatio-Temporal Modeling

I am currently a research assistant professor at the School of Computing and Data Science of HKU, collaborating with Professor Chao Huang. I also work with Professor Ben Kao. My research works focus on large language models and foundation models, graph learning, self-supervised learning, and their applications in recommendation systems and smart cities. I received my Ph.D. and Bachelor degrees in computer science at South China University of Technology (SCUT) adviced by Professor Yong Xu.

Hornors & Awards

  • Worlds's Top 2% Scientists 2024 published by Stanford University.
  • AI 2000 Most Influential Scholars (2022-2023) by Aminer in Data Mining and Information Retrieval:
         Rank 4th (Citation-based), Rank 7th (Contribution-based), Rank 2nd (Rising-Star)
  • One KDD 2024 Most Influential Paper (Rank 10th / 559 Accepted Papers)
  • One WWW 2024 Most Influential Paper (Rank 1st / 405 Accepted Papers)
  • Three WWW 2023 Most Influential Papers (Rank 4th, 5th, 10th / 323 Accepted Papers)
  • One SIGIR 2023 Most Influential Paper (Rank 12th / 165 Accepted Papers)
  • Two KDD 2023 Most Influential Papers (Rank 10th, 11th / 497 Accepted Papers)
  • Two SIGIR 2022 Most Influential Papers (Rank 2nd, 3rd / 161 Accepted Papers)
  • One SIGIR 2021 Most Influential Paper (Rank 13th / 151 Accepted Papers)
  • ACM MM 2024 Best Paper Nomination & Hornorable Mention Award
  • WWW 2023 Best Paper Nomination & Spotlight Paper
  • WSDM 2022 Best Paper Nomination & Top-3 Most Cited Paper
  • WSDM 2023 Top-1 Most Cited Paper
  • ICLR 2023 Spotlight Paper (Top 25%)

    • Selected Publications

    Google scholar citation 4600+, with 2500+ in 2024, h index 34, i-10 index 51

    • [EMNLP'24] "OpenGraph: Towards Open Graph Foundation Models"
      Conference on Empirical Methods in Natural Language Processing
      One of the First Graph Foundation Models, GitHub Star 300+
      L. Xia, B. Kao, and C. Huang
      [pdf] [code]

    • [WWW'23] "Automated Self-Supervised Learning for Recommendation"
      ACM The Web Conference
      Spotlight Paper & Best Paper Nomination (16 / 323), Most Influential Papers of WWW'23 (Rank 10 / 323)
      L. Xia, C. Huang, C.Z. Huang, K. Lin, T. Yu and B. Kao
      [pdf] [code]

    • [WWW'23] "Graph-less Collaborative Filtering"
      ACM The Web Conference
      Top 10% Most-Cited Papers of WWW'2023
      L. Xia, C. Huang, J. Shi and Y. Xu
      [pdf] [code]

    • [ICDE'23] "Disentangled Graph Social Recommendation"
      IEEE International Conference on Data Engineering
      Top 8% Most-Cited Papers of ICDE'2023
      L. Xia, Y. Shao, C. Huang, Y. Xu, H. Xu and J. Pei
      [pdf] [code]

    • [SIGIR'22] "Hypergraph Contrastive Collaborative Filtering"
      International ACM SIGIR Conference on Research and Development in Information Retrieval
      Most Influential Papers of SIGIR'22 (Rank 3 / 161), Citation 360+
      L. Xia, C. Huang, Y. Xu, J. Zhao, D. Yin and J. Huang
      [pdf] [code]

    • [KDD'22] "Self-Supervised Hypergraph Transformer for Recommender Systems"
      ACM SIGKDD Conference on Knowledge Discovery and Data Mining
      Top 7% Most-Cited Papers of KDD'2022, Citation 110+
      L. Xia, C. Huang and C. Zhang
      [pdf] [code]

    • [TKDE'22] "Multi-Behavior Sequential Recommendation with Temporal Graph Transformer"
      IEEE Transactions on Data Engineering
      Top 13% Most-Cited Papers of TKDE'2022
      L. Xia, C. Huang, Y. Xu and J. Pei
      [pdf] [code]

    • [SIGIR'21] "Graph Meta Network for Multi-Behavior Recommendation"
      International ACM SIGIR Conference on Research and Development in Information Retrieval
      Most Influential Papers of SIGIR'21 (Rank 13 / 151), Citation 200+
      L. Xia, Y. Xu, C. Huang, P. Dai and L. Bo
      [pdf] [code]

    • [AAAI'21] "Knowledge-Enhanced Hierarchical Graph Transformer Network for Multi-Behavior Recommendation"
      AAAI Conference on Artificial Intelligence
      Top 6% Most-Cited Papers of AAAI'2021 (among 1692 Accepted Papers), Citation 210+
      L. Xia, C. Huang, Y. Xu, P. Dai, M. Lu and L. Bo
      [pdf] [code]

    • [IJCAI'21] "Spatial-Temporal Sequential Hypergraph Network for Crime Prediction with Dynamic Multiplex Relation Learning"
      International Joint Conference on Artificial Intelligence
      Top 5% Most-Cited Papers of IJCAI'2021 (among 587 Accepted Papers)
      L. Xia, C. Huang, Y. Xu, P. Dai, L. Bo, X. Zhang, T. Chen
      [pdf] [code]

    • [ICDE'21] "Multi-Behavior Enhanced Recommendation with Cross-Interaction Collaborative Relation Modeling"
      IEEE International Conference on Data Engineering
      Top 6% Most-Cited Papers of ICDE'2021
      L. Xia, C. Huang, Y. Xu, P. Dai, M. Lu and L. Bo
      [pdf] [code]

    • [SIGIR'20] "Multiplex Behavioral Relation Learning for Recommendation via Memory Augmented Transformer Network"
      International ACM SIGIR Conference on Research and Development in Information Retrieval
      Top 12% Most-Cited Papers of SIGIR'2020, Citation 120+
      L. Xia, C. Huang, Y. Xu, P. Dai and L. Bo
      [pdf] [code]

    • [Under review] "LightRAG: Simple and Fast Retrieval-Augmented Generation"
      Z. Guo, L. Xia, Y. Yu, T. Ao, and C. Huang
      Receives over 12k GitHub Stars
      [pdf] [code]

    • [KDD'24] "UrbanGPT: Spatio-Temporal Large Language Models"
      ACM SIGKDD Conference on Knowledge Discovery and Data Mining
      Z. Li, L. Xia, J. Tang, Y. Xu, L. Xia, D. Yin, and C. Huang
      Most Influential Papers of KDD'24 (Rank 10 / 559)
      [pdf] [code]

    • [MM'24] "DiffMM: Multi-Modal Diffusion Model for Recommendation"
      ACM International Conference on Multimedia
      Y. Jiang, L. Xia, W. Wei, D. Luo, K. Lin, and C. Huang
      Best Paper Nomination & Hornorable Mention Award
      [pdf] [code]

    • [WWW'24] "RLMRec: Representation Learning with Large Language Models for Recommendation"
      ACM The Web Conference
      X. Ren, W. Wei, L. Xia, L. Su, S. Cheng, J. Wang, D. Yin and C. Huang
      Most Influential Papers of WWW'24 (Rank 1 / 405)
      [pdf] [code]

    • [SIGIR'23] "Disentangled Contrastive Collaborative Filtering"
      International ACM SIGIR Conference on Research and Development in Information Retrieval
      X. Ren, L. Xia, J. Zhao, D. Yin and C. Huang
      Most Influential Papers of SIGIR'23 (Rank 12 / 165)
      [pdf] [code]

    • [WWW'23] "Multi-Modal Self-Supervised Learning for Recommendation"
      ACM The Web Conference
      W. Wei, C. Huang, L. Xia and C. Zhang
      Most Influential Papers of WWW'23 (Rank 4 / 323)
      [pdf] [code]

    • [WWW'23] "Debiased Contrastive Learning for Sequential Recommendation"
      ACM The Web Conference
      Y. Yang, C. Huang, L. Xia, C.Z. Huang, D. Luo and K. Lin
      Most-Influential Papers of WWW'23 (Rank 5 / 323)
      [pdf] [code]

    • [ICLR'23] "LightGCL: Simple Yet Effective Graph Contrastive Learning for Recommendation"
      International Conference on Learning Representations
      Selected as Spotlight Paper (Top 25%)
      X. Cai, C. Huang, L. Xia and X. Ren
      [pdf] [code]

    • [KDD'23] "Knowledge Graph Self-Supervised Rationalization for Recommendation"
      ACM SIGKDD Conference on Knowledge Discovery and Data Mining
      Y. Yang, C. Huang, L. Xia, and C. Huang
      Most Influential Papers of KDD'23 (Rank 11 / 497)
      [pdf] [code]

    • [KDD'23] "Adaptive Graph Contrastive Learning for Recommendation"
      ACM SIGKDD Conference on Knowledge Discovery and Data Mining
      Y. Jiang, C. Huang, and L. Xia
      Most Influential Papers of KDD'23 (Rank 10 / 497)
      [pdf] [code]

    • [WSDM'23] "Heterogeneous Graph Contrastive Learning for Recommendation"
      ACM International Conference on Web Search and Data Mining
      M. Chen, C. Huang, L. Xia, W. Wei, Y. Xu and R. Luo
      Most Cited Paper of WSDM'23 (Rank 1 / 123)
      [pdf] [code]

    • [SIGIR'22] "Knowledge Graph Contrastive Learning for Recommendation"
      International ACM SIGIR Conference on Research and Development in Information Retrieval
      Most Influential Papers of SIGIR'22 (Ranked 2 / 161)
      Y. Yang, C. Huang, L. Xia and C. Li
      [pdf] [code]

    • [WSDM'22] "Contrastive Meta Learning with Behavior Multiplicity for Recommendation"
      ACM International Conference on Web Search and Data Mining
      Best Paper Nomination & Most Cited Papers of WSDM'22 (Rank 3 / 159)
      W. Wei, C. Huang, L. Xia, Y. Xu, J. Zhao, D. Yin
      [pdf] [code]

    Education Background

    • South China University of Technology
      09/2017-12/2021
      Ph.D., at Department of Computer Science and Technology
    • South China University of Technology
      09/2013-06/2017
      Bachelor, Majored in Computer Science and Technology (English-Taught Class)

    Work Experience

    • The University of Hong Kong
      11/2024-Present
      Research Assistant Professor, at School of Computing and Data Science
    • The University of Hong Kong
      04/2022-Present
      Postdoctoral Fellow, at Department of Computer Science and Institute of Data Science
    • Microsoft Research Asia
      09/2016-05/2017
      Research Development Intern, at Innovation Engineering Group

    Services

    Program Commitee Member of Conferences:
    • In year 2024: SIGIR, KDD, WWW, AAAI, NeurIPS, ICLR, CIKM, SDM
    • In year 2023: SIGIR, KDD, WWW, AAAI, NeurIPS
    • In year 2022: KDD, AAAI
    Reviewer for Journals:
    • IEEE Transactions: TKDE, TBD, TNNLS, TLT, TNSE
    • ACM Transactions: TOIS, TWEB, TKDD, TIST
    • APIN, DMKD, Neural Networks, Information Science

    Presentations

    • 08/2024  Tutorial on Large Language Models for Graphs at KDD'2024
    • 05/2024  Tutorial on Large Language Models for Graphs at WWW'2024
    • 10/2022  Tutorial on Self-Supervised Learning for Recommendation at CIKM'2022
    • Research paper presentations at international conferences:
      SIGIR'2023, WWW'2023, ICDE'2023, CIKM'2023, SIGIR'2022, SIGIR'2021, ICDE'2021, SIGIR'2020