Haibo Xiu

About

My name is Haibo Xiu (修海博), a final-year Ph.D. candidate in computer science at Duke University, advised by Professor Jun Yang and Sudeepa Roy. My research focuses on making data systems intelligent, robust, efficient, and explainable. I enjoy taking ideas from research to practice and building production-minded systems. I was an SWE intern at Google (2025), and completed two Research Scientist internships at TikTok (2023 and 2024).

Actively seeking research/industry collaboration focused on LLM/agent for systems and data platforms.

News

• Sep 2025 Four papers (veDB-HTAP, PARQO, PAR2QO, and Hint-QPT) will be presented at VLDB 2025. Many thanks to all collaborators. See you in London! 🇬🇧🎡💂🏼‍♀️
• May 2025 One paper accepted to EDBT 2026 (Tampere, Finland).
• May 2025 Joined Google as a Software Engineering intern.
• May 2024 Joined TikTok as a Research Scientist intern (returned).
• May 2024🎓 Officially became a Ph.D. candidate—many thanks to my committee members for their guidance.

Selected Publications

1. Query Performance Explanation through Large Language Model for HTAP Systems.
   Haibo Xiu, Li Zhang, Tieying Zhang, Jun Yang, and Jianjun Chen.
   EDBT 2026.
2. veDB-HTAP: Highly Integrated, Efficient and Adaptive HTAP System.
   Jianjun Chen, et al. (Infrastructure system lab, ByteDance)
   PVLDB 2025 — Industrial track.
3. PAR2QO: Parametric Penalty-Aware Robust Query Optimization.
   Haibo Xiu, Yang Li, Qianyu Yang, Pankaj K. Agarwal, Jun Yang.
   PVLDB 2025.
4. Hint-QPT: Hints for Robust Query Performance Tuning.
   Haibo Xiu, Yang Li, Qianyu Yang, Weihang Guo, Yuxi Liu, Pankaj K. Agarwal, Sudeepa Roy, Jun Yang.
   PVLDB 2025 — Demo track.
5. PARQO: Penalty-Aware Robust Plan Selection in Query Optimization.
   Haibo Xiu, Pankaj K. Agarwal, Jun Yang.
   PVLDB 2024.
6. CrypQ: A Database Benchmark Based on Dynamic, Ever-Evolving Ethereum Data.
   Vincent Capol, Yuxi Liu, Haibo Xiu, and Jun Yang.
   TPCTC at PVLDB 2024.
7. Selectivity Functions of Range Queries are Learnable.
   Xiao Hu, Yuxi Liu, Haibo Xiu, Pankaj Agarwal, Debmalya Panigrahi, Sudeepa Roy, and Jun Yang.
   SIGMOD 2022.

Invited Talks

• Revisiting Robust Query Optimization — F1 Query, Google (08/05/2025, Sunnyvale, CA)
• Revisiting Robust Query Optimization — Infrastructure System Lab, TikTok (10/18/2024, San Jose, CA)