Publications

2026

  1. Under Review
    A Graph Signal Processing Perspective on Numerical Sequence Representations in LLM In-Context Learning
    Jiajun Bao, Zihao Qi, Toni J.B. Liu, Gurbir Arora, Raphaël Sarfati, Nicolas Boullé, and Christopher J. Earls
    arXiv preprint, 2026
    arXiv
  2. Under Review
    The Geometry of Ignorance: LLMs Know When to Temper Bayesian Priors
    Toni J.B. Liu, Jiajun Bao, Yizhou Liu, Gurbir Arora, Nicolas Boullé, Raphaël Sarfati, and Christopher J. Earls
    arXiv preprint, 2026
    arXiv
  3. Under Review
    Causal and Interpretable Structures in LLM Compositional Tasks
    Gurbir Arora, Toni J.B. Liu, Jiajun Bao, Raphaël Sarfati, and Christopher J. Earls
    arXiv preprint, 2026
    arXiv
  4. ICLR 26 Workshop IOP MLST
    Text-Trained LLMs Can Zero-Shot Extrapolate PDE Dynamics, Revealing a Three-Stage In-Context Learning Mechanism
    Jiajun Bao, Nicolas Boullé, Toni J.B. Liu, Raphaël Sarfati, and Christopher J. Earls
    ICLR 2026 Workshop on AI & PDE – Oral (5.4% oral acceptance rate), 2026
    Invited as a selected workshop contribution and published as part of the AI & PDE Focus Collection in IOP Machine Learning: Science and Technology, 2026.
    arXiv Journal Workshop Code

2024

  1. SIAM/ASA JUQ
    Ensemble Kalman Filters with Resampling
    Omar Al-Ghattas, Jiajun Bao, and Daniel Sanz-Alonso
    SIAM/ASA Journal on Uncertainty Quantification, 2024
    arXiv Journal Code