← Search

Zhongtian Sun

2 accepted papers

2026

Origo: Physically Interpretable Multi-Physics PDE Pre-training through Neural Operator Splitting

ICML 2026poster

Partial Differential Equations (PDEs) play a fundamental role in scientific computing, and recent efforts have sought to extend the success of foundation models to PDE solving. However, multi-physics PDE pre-training faces the unique challenge of disentangling dynamic heterogeneity to learn universa…

Cited by 0SourceScholar
2026

TopAdapter: Topology-Aware Prompt Tuning for Efficient Point Cloud Understanding

ICML 2026poster

Point cloud data, with its inherent geometric and topological structures, plays a critical role in 3D vision tasks. However, existing parameter-efficient fine-tuning (PEFT) methods predominantly focus on input token prompting, overlooking the intrinsic geometric information. To address this limitati…

Cited by 0SourceScholar