collaborators

7 papers

cs.LG2026

SOAP-Bubbles: Structured Weight Uncertainty for Neural Networks

Adrian Robert Minut, Nico Daheim, Marco Miani +3

Structured weight-uncertainty can improve many aspects of deep learning, but it remains costly to estimate and difficult to implement. Here, we show that these issues can be addres…

cs.LG2026

Steering Vectors are an Adversarial Attack Surface

Abzal Aidakhmetov, Donato Crisostomi, Tommaso Mencattini +3

Activation steering has become a popular way to control Large Language Model (LLM) behavior without fine-tuning. Since the technique is plug-and-play, users share datasets and prec…

cs.CV2026

Zero-Shot Quantization via Weight-Space Arithmetic

Daniele Solombrino, Antonio Andrea Gargiulo, Alessandro Zirilli +3

We show that robustness to post-training quantization (PTQ) is a transferable direction in weight space. We call this direction the quantization vector: extracted from a donor task…

cs.CL2026

Multi-objective Evolutionary Merging Enables Efficient Reasoning Models

Mario Iacobelli, Adrian Robert Minut, Tommaso Mencattini +5

Reasoning models achieve strong performance on complex problems by leveraging long chains of thought, but this deliberate reasoning incurs substantial inference-time cost. The Long…

cs.AI2026

Spilled Energy in Large Language Models

Adrian Robert Minut, Hazem Dewidar, Iacopo Masi

We reinterpret the final Large Language Model (LLM) softmax classifier as an Energy-Based Model (EBM), decomposing the sequence-to-sequence probability chain into multiple interact…

cs.LG2025

Mergenetic: a Simple Evolutionary Model Merging Library

Adrian Robert Minut, Tommaso Mencattini, Andrea Santilli +2

Model merging allows combining the capabilities of existing models into a new one - post hoc, without additional training. This has made it increasingly popular thanks to its low c…