Tim Z. Xiao

PhD student at University of Tübingen & International Max Planck Research School for Intelligent Systems

prof_pic.jpg

Hello! I am Tim, also known as Zhenzhong Xiao (肖镇中 in Chinese). Currently, I am doing my PhD in machine learning at the University of Tübingen with Robert Bamler, and as a visiting PhD student at MPI-IS Tübingen with Weiyang Liu and Bernhard Schölkopf.

I am also an organiser of FridayTalks@Tübingen, a bi-weekly AI research seminar aiming to provide a place for the Tübingen AI research community to exchange ideas.

Previously, I did a MRes Computational Statistics and Machine Learning at UCL with David Barber and a MSc in Computer Science at the University of Oxford with Yarin Gal. Prior to that I studied computer science at the University of Manchester for my undergraduate. At the time, I also had a one year industrial placement with Morgan Stanley, during which I worked with a global team to create tools for managing and monitoring the cloud platform with a huge number of servers located around the world.

Just like many computer scientists, I have faith in computing. Although computation will not solve all problems for you, I believe it will certainly take you closer to the answers.

news

Feb 17, 2025 I am honoured to be invited to give the scholar scientific talk at this year’s IMPRS-IS Interview Symposium! The topic is Verbalized Machine Learning, a similar talk can be found here.
Nov 25, 2024 I gave an invited lecture at the University of Michigan on the topic of Verbalized Machine Learning.
Sep 3, 2024 I have been selected by G-Research as one of the Auguest 2024 Grant Winners.

selected publications

  1. vrs_preview.png
    Flipping Against All Odds: Reducing LLM Coin Flip Bias via Verbalized Rejection Sampling
    Tim Z. Xiao, Johannes Zenn, Zhen Liu, Weiyang Liu, Robert Bamler, and Bernhard Schölkopf
    arXiv preprint arXiv:2506.09998, 2025
  2. vml_preview.png
    Verbalized Machine Learning: Revisiting Machine Learning with Language Models
    Tim Z. Xiao, Robert Bamler, Bernhard Schölkopf, and Weiyang Liu
    In Transactions on Machine Learning Research (TMLR), 2025
    ICML 2024 Workshop on In-context Learning
    ICML 2024 Workshop on LLMs and Cognition
  3. dmaapx.png
    A Note on Generalization in Variational Autoencoders: How Effective Is Synthetic Data and Overparameterization?
    Tim Z. Xiao*, Johannes Zenn*, and Robert Bamler
    In Transactions on Machine Learning Research (TMLR), 2024
  4. unireps_rebasin.png
    A Compact Representation for Bayesian Neural Networks By Removing Permutation Symmetry
    Tim Z. Xiao, Weiyang Liu, and Robert Bamler
    arXiv preprint arXiv:2401.00611, 2023
    NeurIPS 2023 Workshop on Unifying Representations in Neural Models
  5. svhn_remix.png
    The SVHN Dataset Is Deceptive for Probabilistic Generative Models Due to a Distribution Mismatch
    Tim Z. Xiao*, Johannes Zenn*, and Robert Bamler
    arXiv preprint arXiv:2312.02168, 2023
    NeurIPS 2023 Workshop on Distribution Shifts
  6. hit_3d.png
    Trading Information between Latents in Hierarchical Variational Autoencoders
    Tim Z. Xiao, and Robert Bamler
    In International Conference on Learning Representations (ICLR), 2023
  7. uncertainty-transformer.png
    Wat zei je? Detecting Out-of-Distribution Translations with Variational Transformers
    Tim Z. Xiao, Aidan N. Gomez, and Yarin Gal
    arXiv preprint arXiv:2006.08344, 2020
    Spotlight talkNeurIPS 2019 Workshop on Bayesian Deep Learning