Xianghong Xu

[Email] [Google Scholar] [ORCID]

About

I currently work at Infrastructure System Lab of ByteDance, focusing on artificial intelligence for data systems.
My research interest and experience involve Information Retrieval (IR), Recommender Systems (RS), Pre-trained Language Models (PLMs), AI for Databases (AI4DB), etc.

Working Experience

Research Engineer

June 2023 - Present

Infrastructure System Lab, ByteDance

Focusing on integrating advanced AI models—including Pre-trained Language Models (PLMs), Large Language Models (LLMs), and Learning-To-Rank (LTR) techniques—into SQL optimization tasks to reduce SQL execution latency. Authored/co-authored 4 papers and contributed to several patent applications.

Education

Tsinghua University

August 2020 - June 2023

Master of Computer Technology

Specialized in Information Retrieval (IR), Recommender Systems (RS), and Natural Language Processing (NLP). Authored/co-authored 8 papers and contributed to 4 patent applications.

Xi'an Jiaotong University

August 2016 - June 2020

Bachelor of Computer Science and Technology

Publications

* indicates equal contributions.
[Order by year] [Order by authorship] [Group by topic]
  1. VIDEX: A Disaggregated and Extensible Virtual Index for the Cloud and AI Era
    Rong Kang, Shuai Wang, Tieying Zhang, Xianghong Xu, Linhui Xu, Zhimin Liang, Lei Zhang, Rui Shi, Jianjun Chen
    VLDB 2025 Demo
    [paper]
  2. PLM4NDV: Minimizing Data Access for Number of Distinct Values Estimation with Pre-trained Language Models
    Xianghong Xu, Xiao He, Tieying Zhang, Lei Zhang, Rui Shi, Jianjun Chen
    SIGMOD 2025
    [paper]
  3. AdaNDV: Adaptive Number of Distinct Value Estimation via Learning to Select and Fuse Estimators
    Xianghong Xu, Tieying Zhang, Xiao He, Haoyang Li, Rong Kang, Shuai Wang, Linhui Xu, Zhimin Liang, Shangyu Luo, Lei Zhang, Jianjun Chen
    VLDB 2025
    [paper]
  4. COOOL: A Learning-To-Rank Approach for SQL Hint Recommendations
    Xianghong Xu, Zhibing Zhao, Tieying Zhang, Rong Kang, Luming Sun, Jianjun Chen
    VLDB 2023 Workshop
    [paper]
  5. Self-supervised Bidirectional Prompt Tuning for Entity-enhanced Pre-trained Language Model
    Jiaxin Zou, Xianghong Xu, Jiawei Hou, Qiang Yan, Hai-Tao Zheng
    IJCNN 2023
    [paper]
  6. Rethinking Temporal Information in Session-Based Recommendation: A Position-Agnostic Approach
    Xianghong Xu*, Kai Ouyang*, Jiaxin Zou, Hai-Tao Zheng, Wenqiang Liu, Dongxiao Huang, Bei Wu
    ECAI 2023
    [paper]
  7. Mining Interest Trends and Adaptively Assigning Sample Weight for Session-based Recommendation
    Kai Ouyang*, Xianghong Xu*, Miaoxin Chen, Zuotong Xie, Hai-Tao Zheng, Shuangyong Song, Yu Zhao
    SIGIR 2023 Short
    [paper]
  8. Modeling Global-Local Subtopic Distribution with Hypergraph to Diversify Search Results
    Kai Ouyang*, Xianghong Xu*, Zuotong Xie, Hai-Tao Zheng, Yanxiong Lu
    IJCNN 2023
    [paper]
  9. Social-aware Sparse Attention Network for Session-based Social Recommendation
    Kai Ouyang*, Xianghong Xu*, Chen Tang, Wang Chen, Hai-Tao Zheng
    EMNLP 2022 Findings
    [paper]
  10. Modeling Latent Autocorrelation for Session-based Recommendation
    Xianghong Xu*, Kai Ouyang*, Liuyin Wang, Jiaxin Zou, Yanxiong Lu, Hai-Tao Zheng, Hong-Gee Kim
    CIKM 2022 Short
    [paper]
  11. Self-Supervised Dual-Channel Attentive Network for Session-based Social Recommendation
    Liuyin Wang*, Xianghong Xu*, Kai Ouyang, Huanzhong Duan, Yanxiong Lu, Hai-Tao Zheng
    ICDE 2022
    [paper]
  12. Diversify Search Results Through Graph Attentive Document Interaction
    Xianghong Xu*, Kai Ouyang*, Yin Zheng, Yanxiong Lu, Hai-Tao Zheng, Hong-Gee Kim
    DASFAA 2022
    [paper]