Guo-Wei Wei

Geometric variability in MOF coordination cores, invisible to classical topology, is now quantifiable and predictive using persistent Mayer homology.

Topological spectral representations and latent interaction attention unlock structured, source-target-dependent information exchange for biomolecular learning, achieving SOTA drug discovery performance.

Unlocking superior molecular and materials property prediction by fusing diverse mathematical invariants with topological neural networks, revealing that optimal performance hinges on carefully curated representation-architecture pairings.

Language models can orchestrate scientific discovery by optimizing complex mathematical representations for molecular property prediction, revealing dataset-specific invariant needs.