Haiwen Huang

I am Haiwen Huang (黄海文), a postdoc in Prof. Moritz Hardt’s group at the Max Planck Institute for Intelligent Systems (MPI-IS) in Tübingen. Previously, I completed my PhD at the University of Tübingen, co-supervised by Prof. Andreas Geiger and Dr. Dan Zhang. My current research interests are in building computational foundations for social simulation, from benchmarks to simulation methodologies.

My PhD research focused on improving the generalization of vision and multimodal models, enabling them to perform robustly across diverse tasks and domains. For example, I have developed methods that leverage 3D priors to enhance 2D object detection in GOOD (ICLR 2023) and use self-distillation to upsample features in Vision Foundation Models in LoftUp (ICCV 2025). I have also contributed to building more reliable evaluations of open-vocabulary generalization in RENOVATE (NeurIPS 2024).

Before my PhD, I completed my MSc in CS at University of Oxford, advised by Prof. Yarin Gal, and my undergrad in Mathematics at Peking University, advised by Prof. Bin Dong. I also worked as a researcher at Megvii (Face++) with Xinyu Zhou, studying large-scale annotation and OoD detection.

I am honored to receive the NeurIPS 2024 Top Reviewer Award and the CVPR 2025 Outstanding Reviewer Award.