Official codebase of "Update Your Transformer to the Latest Release: Re-Basin of Task Vectors" - ICML 2025

Utils and methods for weight-space merging.

Repository of the paper "MERGE^3 : efficient evolutionary merging on consumer-grade GPUs" (ICML 2025)

Website for the Gladia Research Group at Sapienza University of Rome, led by Prof. E. Rodolà.

A Python package for analyzing and transforming neural latent spaces.

LaTeX Plugin for Adobe Illustrator

Flexible library for merging large language models (LLMs) via evolutionary optimization (ACL 2025 Demo).

Generic template to bootstrap your Python project.

Standardized DataLoaders for 3D Computer Vision

Task Singular Vectors: Reducing Task Interference in Model Merging. Merge models avoiding task interference through separable models.

Codebase for "C2M3: Cycle-Consistent Multi-Model Merging".

A Python package for analyzing and transforming neural latent spaces.

Few-Shot Graph Classification via distance metric learning.

Canonical Correlation Analysis Zoo: A collection of Regularized, Deep Learning based, Kernel, and Probabilistic methods in a scikit-learn style framework

Editing Models with Task Arithmetic

Relative representations can be leveraged to enable solving tasks regarding "latent communication": from zero-shot model stitching to latent space comparison between diverse settings.

Official Code for DragGAN (SIGGRAPH 2023)

Better Enabled Resources

Generic template to bootstrap your PyTorch project.

Teaching material for the course of Deep Learning and Applied AI, 2nd semester 2020, Sapienza University of Rome