Generative AI

Cosmo3DFlow: Wavelet Flow Matching for Spatial-to-Spectral Compression in Reconstructing the Early Universe

Flow matching with 3D wavelet representations for cosmological reconstruction: up to 46x faster sampling than diffusion models on 128^3 simulations (KDD 2026).

Md Khairul Islam

• Aug 2026 • 1 min read

Flow Matching

ScatterPrism: convergence for generative simulation and inverse problems in particle and nuclear physics

Conditional flow matching for particle simulation and inverse problems, with multi-metric evaluation of physical fidelity beyond training loss.

Zeyu Xia

• Jul 2026 • 1 min read

No results found

Generative AI

Cosmo3DFlow: Wavelet Flow Matching for Spatial-to-Spectral Compression in Reconstructing the Early Universe

ScatterPrism: convergence for generative simulation and inverse problems in particle and nuclear physics