Papers &
preprints.

MA-ProofBench: A Two-Tiered Evaluation of LLMs for Theorem Proving in Mathematical Analysis

Lushi Pu, Weiming Zhang, Xinheng Xie, Zixuan Fu, Bingxiang He, Hongya Lyu, Xin Li, Jie Zhou, Yudong Wang

arXiv:2606.13782

How well can language models prove theorems? A two-tier benchmark for formal theorem proving in mathematical analysis.

2DSig-Detect: A semi-supervised framework for anomaly detection on image data using 2D-signatures

Xinheng Xie, Kureha Yamaguchi, Margaux Leblanc, Simon Malzard, Varun Chhabra, Victoria Nockles, Yue Wu

Pattern Recognition, 180, 114172

Finding the unusual in an image. Bringing 2D-signatures and semi-supervised learning together for anomaly detection.

The publisher’s bibliographic record assigns this article to the December 2026 issue.

NODE-ImgNet: A PDE-informed effective and robust model for image denoising

Xinheng Xie, Yue Wu, Hao Ni, Cuiyu He

Pattern Recognition, 148, 110176

A clearer image, through mathematics. Using differential-equation structure and neural ODEs to learn how to remove noise.

NODE-AdvGAN: Improving the transferability and perceptual similarity of adversarial examples by dynamic-system-driven adversarial generative model

Xinheng Xie, Yue Wu, Cuiyu He

arXiv:2412.03539

Small changes, different answers. Studying model vulnerabilities through a dynamical approach to adversarial image generation.

On approximation of solutions of stochastic delay differential equations via randomized Euler scheme

Paweł Przybyłowicz, Yue Wu, Xinheng Xie

Applied Numerical Mathematics, 197, 143–163

Reliable numerical approximations for systems influenced by randomness and their past states.

A Randomized Runge-Kutta Method for time-irregular delay differential equations

Fabio V. Difonzo, Paweł Przybyłowicz, Yue Wu, Xinheng Xie

arXiv:2401.11658

A randomized numerical method for delay differential equations with irregular dependence on time.