3 papers
astro-ph.IM2026
The indiscriminate adoption of AI threatens the foundations of academia
Roberto Trotta
Artificial intelligence offers much promise, but its use in scientific research should be restrained so that the primary aim of academia -- advancing knowledge for humans -- is saf…
astro-ph.CO2026
Cosmo-FOLD: Fast generation and upscaling of field-level cosmological maps with overlap latent diffusion
Satvik Mishra, Roberto Trotta, Matteo Viel
We demonstrate the capabilities of probabilistic diffusion models to reduce dramatically the computational cost of expensive hydrodynamical simulations to study the relationship be…
astro-ph.CO2025
JERALD: high-fidelity dark matter, stellar mass and neutral hydrogen maps from fast N-body simulations
Mauro Rigo, Roberto Trotta, Matteo Viel
We present a new code and approach, JERALD -- JAX Enhanced Resolution Approximate Lagrangian Dynamics -- , that improves on and extends the Lagrangian Deep Learning method of Dai &…