Haoxin Liu

Haoxin Liu

Ph.D. student in Computer Science at Georgia Tech focusing on time-series forecasting, graph learning, and large language models.

Haoxin Liu

Haoxin Liu is a Ph.D. student in Computer Science at the Georgia Institute of Technology advised by Prof. B. Aditya Prakash. His research spans time-series forecasting, graph learning, foundation models, and recommender systems, with publications across KDD, NeurIPS, ICML, ACL, NAACL, WWW, SIGIR, and other venues.

  • Research interests: time-series foundation models, multimodal learning, distribution shift robustness, graph neural networks, and recommender systems.
  • Contact: hliu763@gatech.edu
  • Google Scholar · LinkedIn · GitHub
  • 📄 Download CV (PDF)

Education

  • Georgia Institute of Technology, Atlanta, United States
    Ph.D. in Computer Science, Aug. 2023 – Aug. 2028 (expected)
    Advised by Prof. B. Aditya Prakash · GPA: 4.0/4.0
  • Tsinghua University, Beijing, China
    M.S. in Computer Science and Technology, Sep. 2020 – Jul. 2023
    Advised by Prof. Wenwu Zhu and Prof. Peng Cui · GPA: 3.81/4.0
  • Shandong University, Shandong, China
    B.S. in Computer Science and Technology, Sep. 2016 – Jul. 2020
    GPA: 91.50/100 (top 5%)

Publications in Peer-Reviewed Conferences (Time Series; Graph; LLMs; Recommender Systems)

  • [C11] Zhiyuan Zhao, Haoxin Liu, Alexander Rodríguez, B. Aditya Prakash. Performative Time-Series Forecasting. KDD 2025.
  • [C10] Shangqing Xu, Harshavardhan Kamarthi, Haoxin Liu, B. Aditya Prakash. In-context Pre-trained Time-Series Foundation Models adapt to Unseen Tasks. CIKM 2025.
  • [C9] Haoxin Liu, Chenghao Liu, B. Aditya Prakash. A Picture is Worth A Thousand Numbers: Enabling LLMs Reason about Time Series via Visualization. NAACL 2025 Main Conference.
  • [C8] Haoxin Liu, Shangqing Xu, Zhiyuan Zhao, Lingkai Kong, Harshavardhan Kamarthi, Aditya B. Sasanur, Megha Sharma, Jiaming Cui, Qingsong Wen, Chao Zhang, B. Aditya Prakash. Time-MMD: A New Multi-Domain Multimodal Dataset for Time Series Analysis. NeurIPS 2024 (acceptance rate 25.8%).
  • [C7] Haoxin Liu, Harshavardhan Kamarthi, Lingkai Kong, Zhiyuan Zhao, Chao Zhang, B. Aditya Prakash. Time-Series Forecasting for Out-of-Distribution Generalization Using Invariant Learning. ICML 2024 (acceptance rate 27.5%).
  • [C6] Haoxin Liu, Zhiyuan Zhao, Jindong Wang, Harshavardhan Kamarthi, B. Aditya Prakash. LSTPrompt: Large Language Models as Zero-Shot Time Series Forecasters by Long-Short-Term Prompting. ACL 2024 Findings.
  • [C5] Haoxin Liu, Pu Zhao, Si Qin, Yong Shi, Mirror Xu, Qingwei Lin, Dongmei Zhang. HAPENS: Hardness-Personalized Negative Sampling for Implicit Collaborative Filtering. WWW 2023 Industry Track (acceptance rate 19.8%).
  • [C4] Haoxin Liu. LightSGCN: Powering Signed Graph Convolution Network for Link Sign Prediction with Simplified Architecture Design. SIGIR 2022 (sole author, acceptance rate 24.7%).
  • [C3] Haoxin Liu, Ziwei Zhang, Peng Cui, Yafeng Zhang, Qiang Cui, Jiashuo Liu, Wenwu Zhu. Signed Graph Neural Network with Latent Groups. KDD 2021 Research Track (acceptance rate 15.4%).
  • [C3] Bin Jiang, Haoxin Liu, Qingwei Li, Shuhua Cao, Zhaoyu Chen, Liyun Cheng, Meixia Qu. Feature Extraction of SDSS Spectra With Improved CNN. ICCSSE 2019 (First student author, Best Oral Presentation Award).
  • [C2] Xingxuan Zhang, Linjun Zhou, Renzhe Xu, Peng Cui, Zheyan Shen, Haoxin Liu. Towards Unsupervised Domain Generalization. CVPR 2022 (acceptance rate 25.3%).
  • [C1] Qian Yu, Xiangdong Wu, Chen Yang, Zihao Zhao, Haoxin Liu, Jingping Shao. Exploiting Global Behavior Contextual Correlation in Sequential Recommendation. CIKM 2022 DL4SR.

Works Under Review

  • [U6] Guancheng Wan, Lucheng Fu, Haoxin Liu, Yiqiao Jin, Hui Yi Leong, Eric Hanchen Jiang, Hejia Geng, Jinhe Bi, Yunpu Ma, Xiangru Tang, B. Aditya Prakash, Yizhou Sun, Wei Wang. Beyond Magic Words: Sharpness-Aware Prompt Evolving for Robust Large Language Models with TARE.
  • [U5] Haoxin Liu, Harshavardhan Kamarthi, Zhiyuan Zhao, Shangqing Xu, Shiduo Wang, Qingsong Wen, Tom Hartvigsen, Fei Wang, B. Aditya Prakash. How can time series analysis benefit from multiple modalities? A survey and outlook (arXiv).
  • [U4] Haoxin Liu, Zhiyuan Zhao, Shiduo Li, B. Aditya Prakash. Evaluating System 1 vs. 2 Reasoning Approaches for Zero-Shot Time-Series Forecasting: A Benchmark and Insights.
  • [U3] Zhiyuan Zhao, Haoxin Liu, B. Aditya Prakash. Navigating Concept Drift and Temporal Shift: Distribution Shift Generalized Time-Series Forecasting.
  • [U2] Zhiyuan Zhao, Juntong Ni, Haoxin Liu, Shangqing Xu, Wei Jin, B. Aditya Prakash. TIMERECIPE: A Time-Series Forecasting Recipe via Benchmarking Module Level Effectiveness.
  • [U1] Wenjie Du, Jun Wang, Linglong Qian, Yiyuan Yang, Zina Ibrahim, Fanxing Liu, Zepu Wang, Haoxin Liu, Zhiyuan Zhao, Yingjie Zhou, Wenjia Wang, Kaize Ding, Yuxuan Liang, B. Aditya Prakash, Qingsong Wen. TSI-Bench: Benchmarking Time Series Imputation.

Intern Experience

  • Remote Student Researcher, Google Research — Aug. 2025 – Dec. 2025
    Mentored by Dr. Rajat Sen.
    • Scale up multimodal time-series datasets and train large forecasting foundation models.
  • Student Researcher, Google Research — May 2025 – Aug. 2025
    Mentored by Dr. Rajat Sen.
    • Built a Deep Research Agent to automatically construct multimodal time-series datasets and evaluated multiple multimodal forecasting approaches.
  • Research Engineer Intern, Youxuan (Meituan) — Jul. 2021 – Nov. 2021
    Mentored by Dr. Qiang Cui.
    • Focused on recommender systems, published work on modeling negative feedback, and contributed to a patent for real-world deployment.

Selected Awards & Honors

  • “Star of Tomorrow” Award of Excellence — Microsoft Research Asia, 2022.
  • Top Reviewer (Registration Fee Reward) — NeurIPS Program, 2024.