14 papers
Compositional Generalization in Autoregressive Models via Logit Composition
Aakash Kumar, Maria Sofia Bucarelli, Emanuele Natale
Composing autoregressive models remains a core challenge in understanding how large language models can combine behaviors or skills learned across tasks. We introduce a new and pri…
Select, Label, Evaluate: Active Testing in NLP
Antonio Purificato, Maria Sofia Bucarelli, Andrea Bacciu +2
Human annotation cost and time remain significant bottlenecks in Natural Language Processing (NLP), with test data annotation being particularly expensive due to the stringent requ…
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…
The Majority Vote Paradigm Shift: When Popular Meets Optimal
Antonio Purificato, Maria Sofia Bucarelli, Anil Kumar Nelakanti +3
Reliably labelling data typically requires annotations from multiple human workers. However, humans are far from being perfect. Hence, it is a common practice to aggregate labels g…
Same Answer, Different Representations: Hidden instability in VLMs
Farooq Ahmad Wani, Alessandro Suglia, Rohit Saxena +6
The robustness of Vision Language Models (VLMs) is commonly assessed through output-level invariance, implicitly assuming that stable predictions reflect stable multimodal processi…
Energy Guided smoothness to improve Robustness in Graph Classification
Farooq Ahmad Wani, Maria Sofia Bucarelli, Andrea Giuseppe Di Francesco +2
Graph Neural Networks (GNNs) are powerful at solving graph classification tasks, yet applied problems often contain noisy labels. In this work, we study GNN robustness to label noi…