Vittoria Vineis

Filippo Betello, Antonio Purificato, Vittoria Vineis et al.

The environmental impact of Artificial Intelligence (AI) is emerging as a significant global concern, particularly regarding model training. In this paper, we introduce GREEN (Guided Recommendations of Energy-Efficient Networks), a novel, inference-time approach for recommending Pareto-optimal AI model configurations that optimize validation performance and energy consumption across diverse AI domains and tasks. Our approach directly addresses the limitations of current eco-efficient neural architecture search methods, which are often restricted to specific architectures or tasks. Central to this work is EcoTaskSet, a dataset comprising training dynamics from over 1767 experiments across computer vision, natural language processing, and recommendation systems using both widely used and cutting-edge architectures. Leveraging this dataset and a prediction model, our approach demonstrates effectiveness in selecting the best model configuration based on user preferences. Experimental results show that our method successfully identifies energy-efficient configurations while ensuring competitive performance.

Vittoria Vineis, Giuseppe Perelli, Gabriele Tolomei

Conventional automated decision-support systems, often based on supervised learning, focus on predicting outcomes to recommend actions. However, they typically overlook the complexity of multi-actor environments, where diverse and conflicting stakeholder preferences must be balanced. At the same time, participatory AI approaches remain largely context-specific, limiting their broader applicability. To address these gaps, we propose a participatory framework that reframes decision-making as a multi-stakeholder optimization problem, using context-dependent reward functions to represent each actor's preferences. Our modular, model-agnostic framework employs k-fold cross-validation to fine-tune user-provided prediction models and evaluate decision strategies, including compromise functions that mediate stakeholder trade-offs. A synthetic scoring mechanism aggregates user-defined preferences across multiple metrics to rank strategies and select an optimal decision-maker for generating actionable recommendations on new data. Validated on two high-stake real-world case studies, the framework consistently produces stakeholder-aware decisions that outperform purely predictive baselines across multiple metrics, while enhancing the transparency and accountability of AI-supported decision-making.

Vittoria Vineis, Matteo Silvestri, Lorenzo Antonelli et al.

Explainable Artificial Intelligence (XAI) seeks to enhance the transparency and accountability of machine learning systems, yet most methods follow a one-size-fits-all paradigm that neglects user differences in expertise, goals, and cognitive needs. Although Large Language Models can translate technical explanations into natural language, they introduce challenges related to faithfulness and hallucinations. To address these challenges, we present PONTE (Personalized Orchestration for Natural language Trustworthy Explanations), a human-in-the-loop framework for adaptive and reliable XAI narratives. PONTE models personalization as a closed-loop validation and adaptation process rather than prompt engineering. It combines: (i) a low-dimensional preference model capturing stylistic requirements; (ii) a preference-conditioned generator grounded in structured XAI artifacts; and (iii) verification modules enforcing numerical faithfulness, informational completeness, and stylistic alignment, optionally supported by retrieval-grounded argumentation. User feedback iteratively updates the preference state, enabling quick personalization. Automatic and human evaluations across healthcare and finance domains show that the verification-refinement loop substantially improves completeness and stylistic alignment over validation-free generation. Human studies further confirm strong agreement between intended preference vectors and perceived style, robustness to generation stochasticity, and consistently positive quality assessments.