Research Scientist at Meta
I am Zhibing Zhao, a Ph.D. in Computer Science based in Bellevue, WA. My work spans ads prediction, learning to rank, recommendation systems, natural language processing, time series, query optimization, statistics, and game theory.
Research direction
My research combines practical industrial ML with foundations in preference learning and social choice, connecting production-scale ranking problems with rigorous models of uncertainty, utility, and choice.
Focus areas
At Meta, I work on conversion prediction with click intent signals, plus end-to-end model building, training, and launch workflows that move measurable global ads metrics.
I design ranking and recommendation models that balance relevance gains with the constraints of real product pipelines, where latency, maintainability, and feature quality all matter.
I work on models that help infrastructure choose better plans and forecasts, from SQL hint recommendation to adaptive storage-usage prediction.
Publications
Xianghong Xu, Zhibing Zhao, Tieying Zhang, Rong Kang, Luming Sun, and Jianjun Chen. AIDB 2023.
Read paperLuming Sun, Shijin Gong, Tieying Zhang, Fuxin Jiang, Zhibing Zhao, Jianjun Chen, and Xinyu Zhang. ICDE 2023 Industry and Applications Track.
Read paperZhibing Zhao, Ao Liu, and Lirong Xia. Proceedings of the 31st International Joint Conference on Artificial Intelligence.
Read paperZhibing Zhao and Lirong Xia. Proceedings of the 33rd Conference on Neural Information Processing Systems.
Read paperExperience
Machine learning research for ads prediction and product-facing ranking systems.
SQL query optimization with learning-to-rank and storage usage forecasting.
Hybrid ranking models for search and page recommendation.
Preference learning and aggregation from rank data.
Education
Contact