I am Shengwei Xu. I received my Ph.D. from the University of Michigan’s School of Information, where I was advised by Prof. Grant Schoenebeck. I am joining Google in June 2026. I received my B.S. in Computer Science from Peking University in 2021, advised by Prof. Yuqing Kong.

Research Interests

My dissertation, Aligned Information Elicitation for Text, studies how to design elicitation and evaluation methods for text that are aligned with the informativeness. The central idea is that statistical alignment is not enough on its own: evaluation scores should correlate with human ratings, but they also need strategic alignment, so that the scoring rule discourages untruthful or low-quality reports even when those reports are out-of-distribution.

More broadly, my research investigates incentives and strategic behavior in interactions between agents (human and AI) and algorithm, including mechanism design, information elicitation, and their intersection with Large Language Models.

Publications

• Ad Insertion in LLM-Generated Responses.
  In the Workshop on Frontiers of Online Advertising at the 26th ACM Conference on Economics and Computation (EC’25)
  Shengwei Xu*, Zhaohua Chen*, Xiaotie Deng, Zhiyi Huang, Grant Schoenebeck.
  [paper]

• Stochastically Dominant Peer Prediction.
  In Advances in Neural Information Processing Systems (NeurIPS 2025)
  Yichi Zhang, Shengwei Xu, David Pennock, Grant Schoenebeck.
  [paper]

• Benchmarking LLMs’ Judgments with No Gold Standard.
  In Proceedings of the Thirteenth International Conference on Learning Representations (ICLR 2025)
  Shengwei Xu*, Yuxuan Lu*, Grant Schoenebeck, Yuqing Kong.
  [paper]

• Eliciting Informative Text Evaluations with Large Language Models.
  In Proceedings of the 25th ACM Conference on Economics and Computation (EC’24)
  Yuxuan Lu*, Shengwei Xu*, Yichi Zhang, Yuqing Kong, Grant Schoenebeck.
  [paper] [oral]

• Spot Check Equivalence: An Interpretable Metric for Information Elicitation Mechanisms.
  In Proceedings of the ACM Web Conference 2024 (WWW’24)
  Shengwei Xu*, Yichi Zhang, Paul Resnick, Grant Schoenebeck.
  [paper] [oral]

• BONUS! Maximizing Surprise.
  In Proceedings of the ACM Web Conference 2022 (WWW’22)
  Zhihuan Huang, Yuqing Kong, Tracy Xiao Liu, Grant Schoenebeck, Shengwei Xu.
  [paper]

• SURPRISE! and When to Schedule It.
  In Proceedings of the 30th International Joint Conference on Artificial Intelligence (IJCAI-21)
  Zhihuan Huang*, Shengwei Xu*, You Shan, Yuxuan Lu, Yuqing Kong, Tracy Xiao Liu, Grant Schoenebeck.
  [paper]

Tutorials

• Information Elicitation Meets Large Language Models: Progress, Opportunities, and Challenges. Tutorial in the 20th Conference on Web and Internet Economics (WINE 2024) [Link]

Experiences

Education

• University of Michigan, School of Information, Ph.D., 2021 - 2026.
• Peking University, School of EECS, B.S. in Computer Science, Summa Cum Laude, 2017 - 2021.

Industry

• Google, Incoming Software Engineer, 2026.

• Google, Software Engineering Intern (PhD), May 2025 - Aug 2025.

  • Research and develop a policy classification model for ad safety that utilizes advertiser temporal behaviors.
  • Implement a temporal transformer neural network for multi-modality advertiser data with TensorFlow.

Teaching

• Graduate Student Instructor
  University of Michigan
  • Mathematical Foundations for Applied Data Science (Winter 2024)
  • Applied Machine Learning (Fall 2023)
  • Data-Oriented Programming (Fall 2022)
• Teaching Assistant
  Peking University
  • Discrete Mathematics and Structures (I) and (II) (Spring 2020 and Fall 2020)