selected

MASS: MoErging through Adaptive Subspace Selection
Model merging has recently emerged as a lightweight alternative to ensembling, combining multiple fine-tuned models into a single set …
Donato Crisostomi, Alessandro Zirilli, Antonio Andrea Gargiulo, Maria Sofia Bucarelli, Simone Scardapane, Fabrizio Silvestri, Iacopo Masi, Emanuele Rodolà
Cite arXiv GitHub
MASS: MoErging through Adaptive Subspace Selection
MERGE³: Efficient Evolutionary Merging on Consumer-grade GPUs
Evolutionary model merging enables the creation of high-performing multi-task models but remains computationally prohibitive for …
Tommaso Mencattini, Adrian Robert Minut, Donato Crisostomi, Andrea Santilli, Emanuele Rodolà
Cite arXiv GitHub Tweeprint
MERGE³: Efficient Evolutionary Merging on Consumer-grade GPUs
Task Singular Vectors: Reducing Task Interference in Model Merging
Task Arithmetic has emerged as a simple yet effective method to merge models without additional training. However, by treating entire …
Antonio Andrea Gargiulo, Donato Crisostomi, Maria Sofia Bucarelli, Simone Scardapane, Emanuele Rodolà
Cite arXiv GitHub Tweeprint
Task Singular Vectors: Reducing Task Interference in Model Merging
C²M³: Cycle-Consistent Multi-Model Merging
In this paper, we present a novel data-free method for merging neural networks in weight space. Differently from most existing works, …
Donato Crisostomi, Marco Fumero, Daniele Baieri, Florian Bernard, Emanuele Rodolà
Cite arXiv GitHub Tweeprint
C²M³: Cycle-Consistent Multi-Model Merging
Metric Based Few-Shot Graph Classification
Few-shot graph classification is a novel yet promising emerging research field that still lacks the soundness of well-established …
Donato Crisostomi, Simone Antonelli, Valentino Maiorca, Luca Moschella, Riccardo Marin, Emanuele Rodolà
Cite URL PDF GitHub
Metric Based Few-Shot Graph Classification