CV

Education

• B.S. in Computational Mathematics, Peking University, 2019
• M.S. in Computer Science, University of Oxford, 2021
• Ph.D in Computer Science, University of Tübingen, 2026 (expected)

Work experience

• July 2025 - Oct. 2025: Student Researcher, Google Zurich
  • Research in improving multimodal reasoning abilities of multimodal LLMs
• Feb. 2019 - Sept. 2020: ML Researcher, Data Research Group, Megvii Technology Inc
  • Research projects: Large-scale data annotation, out-of-distribution detection, few-shot learning
  • Group leader: Xinyu Zhou

Selected Publications

Nostalgic Adam: Weighing more of the past gradients when designing the adaptive learning rate

Feature Space Singularity for Out-of-distribution Detection

Can convolutional ResNets approximately preserve input distances? A frequency analysis perspective

Decomposing Representations for Deterministic Uncertainty Estimation