5 papers
FLAGG: Flexible Autoregressive Graph Generation
Samuel Cognolato, Alessandro Sperduti, Luciano Serafini
The Deep Graph Generation's panorama spans two extremes: one-shot and sequential models. The former generates nodes and edges jointly, while the latter samples them autoregressivel…
A City-Scale Dataset of Traffic Flows, Travel Times, and Urban Context
Riccardo Cappi, Massimiliano Luca, Pietro Fontolan +3
We present a multi-source traffic dataset derived from Automatic Vehicle Identification (AVI) recordings in Padua, Italy, spanning from February 2026 to April 2026. The dataset com…
Iterative In-Context Learning to Enhance LLMs Abstract Reasoning: The Case-Study of Algebraic Tasks
Stefano Fioravanti, Matteo Zavatteri, Roberto Confalonieri +4
LLMs face significant challenges in systematic generalization, particularly when dealing with reasoning tasks requiring compositional rules and handling out-of-distribution example…
Discovering Generalizable Governing Equations for Graph Dynamical Systems with Interpretable Neural Networks
Riccardo Cappi, Paolo Frazzetto, Nicolò Navarin +1
The discovery of symbolic governing equations is a central goal in science; yet, it remains challenging particularly for graph dynamical systems, where the network topology further…
LLMs as Data Annotators: How Close Are We to Human Performance
Muhammad Uzair Ul Haq, Davide Rigoni, Alessandro Sperduti
In NLP, fine-tuning LLMs is effective for various applications but requires high-quality annotated data. However, manual annotation of data is labor-intensive, time-consuming, and…