5 papers
LRCC: Generalizing Low-Rank Compression with Conditional Computation
Thomas Vaitses Fontanari, Maximo Eduardo Rulli, Federico Alvetreti +2
Low-rank compression reduces the cost of pretrained language models by replacing linear transformations with low-rank factorizations. However, conventional methods use a fixed rank…
Subliminal Clocks: Latent Time Modelling in Diffusion Language Models
Maximo Eduardo Rulli, Thomas Vaitses Fontanari, Simone Petruzzi +9
Diffusion Language Models (DLMs) have recently emerged as a promising alternative to autoregressive models. Unlike standard diffusion-based approaches, DLMs are not explicitly cond…
Attention Sinks in Diffusion Language Models
Maximo Eduardo Rulli, Simone Petruzzi, Edoardo Michielon +3
Masked Diffusion Language Models (DLMs) have recently emerged as a promising alternative to traditional Autoregressive Models (ARMs). DLMs employ transformer encoders with bidirect…
MASS: MoErging through Adaptive Subspace Selection
Donato Crisostomi, Alessandro Zirilli, Antonio Andrea Gargiulo +5
Model merging has recently emerged as a lightweight alternative to ensembling, combining multiple fine-tuned models into a single set of parameters with no additional training over…
Task Singular Vectors: Reducing Task Interference in Model Merging
Antonio Andrea Gargiulo, Donato Crisostomi, Maria Sofia Bucarelli +3
Task Arithmetic has emerged as a simple yet effective method to merge models without additional training. However, by treating entire networks as flat parameter vectors, it overloo…