Marco Viviani

Rishabh Upadhyay, Arian Askari, Gabriella Pasi et al.

In this paper, we propose a novel approach to consider multiple dimensions of relevance beyond topicality in cross-encoder re-ranking. On the one hand, current multidimensional retrieval models often use naïve solutions at the re-ranking stage to aggregate multiple relevance scores into an overall one. On the other hand, cross-encoder re-rankers are effective in considering topicality but are not designed to straightforwardly account for other relevance dimensions. To overcome these issues, we envisage enhancing the candidate documents -- which are retrieved by a first-stage lexical retrieval model -- with "relevance statements" related to additional dimensions of relevance and then performing a re-ranking on them with cross-encoders. In particular, here we consider an additional relevance dimension beyond topicality, which is credibility. We test the effectiveness of our solution in the context of the Consumer Health Search task, considering publicly available datasets. Our results show that the proposed approach statistically outperforms both aggregation-based and cross-encoder re-rankers.

Pranav Kasela, Marco Braga, Alessandro Ghiotto et al.

In this paper, we present DIETA, a small, decoder-only Transformer model with 0.5 billion parameters, specifically designed and trained for Italian-English machine translation. We collect and curate a large parallel corpus consisting of approximately 207 million Italian-English sentence pairs across diverse domains, including parliamentary proceedings, legal texts, web-crawled content, subtitles, news, literature and 352 million back-translated data using pretrained models. Additionally, we create and release a new small-scale evaluation set, consisting of 450 sentences, based on 2025 WikiNews articles, enabling assessment of translation quality on contemporary text. Comprehensive evaluations show that DIETA achieves competitive performance on multiple Italian-English benchmarks, consistently ranking in the second quartile of a 32-system leaderboard and outperforming most other sub-3B models on four out of five test suites. The training script, trained models, curated corpus, and newly introduced evaluation set are made publicly available, facilitating further research and development in specialized Italian-English machine translation. https://github.com/pkasela/DIETA-Machine-Translation

Ipek Baris Schlicht, Burcu Sayin, Zhixue Zhao et al.

Equitable access to reliable health information is vital when integrating AI into healthcare. Yet, information quality varies across languages, raising concerns about the reliability and consistency of multilingual Large Language Models (LLMs). We systematically examine cross-lingual disparities in pre-training source and factuality alignment in LLM answers for multilingual healthcare Q&A across English, German, Turkish, Chinese (Mandarin), and Italian. We (i) constructed Multilingual Wiki Health Care (MultiWikiHealthCare), a multilingual dataset from Wikipedia; (ii) analyzed cross-lingual healthcare coverage; (iii) assessed LLM response alignment with these references; and (iv) conducted a case study on factual alignment through the use of contextual information and Retrieval-Augmented Generation (RAG). Our findings reveal substantial cross-lingual disparities in both Wikipedia coverage and LLM factual alignment. Across LLMs, responses align more with English Wikipedia, even when the prompts are non-English. Providing contextual excerpts from non-English Wikipedia at inference time effectively shifts factual alignment toward culturally relevant knowledge. These results highlight practical pathways for building more equitable, multilingual AI systems for healthcare.

Xiancheng Li, Georgios D. Karampatakis, Helen E. Wood et al.

Digital health analytics face critical challenges nowadays. The sophisticated analysis of patient-generated health content, which contains complex emotional and medical contexts, requires scarce domain expertise, while traditional ML approaches are constrained by data shortage and privacy limitations in healthcare settings. Online Health Communities (OHCs) exemplify these challenges with mixed-sentiment posts, clinical terminology, and implicit emotional expressions that demand specialised knowledge for accurate Sentiment Analysis (SA). To address these challenges, this study explores how Large Language Models (LLMs) can integrate expert knowledge through in-context learning for SA, providing a scalable solution for sophisticated health data analysis. Specifically, we develop a structured codebook that systematically encodes expert interpretation guidelines, enabling LLMs to apply domain-specific knowledge through targeted prompting rather than extensive training. Six GPT models validated alongside DeepSeek and LLaMA 3.1 are compared with pre-trained language models (BioBERT variants) and lexicon-based methods, using 400 expert-annotated posts from two OHCs. LLMs achieve superior performance while demonstrating expert-level agreement. This high agreement, with no statistically significant difference from inter-expert agreement levels, suggests knowledge integration beyond surface-level pattern recognition. The consistent performance across diverse LLM models, supported by in-context learning, offers a promising solution for digital health analytics. This approach addresses the critical challenge of expert knowledge shortage in digital health research, enabling real-time, expert-quality analysis for patient monitoring, intervention assessment, and evidence-based health strategies.

Alessandro Raganato, Rafael Peñaloza, Marco Viviani et al.

Large Language Models (LLMs) have shown remarkable capabilities in manipulating natural language across multiple applications, but their ability to handle simple reasoning tasks is often questioned. In this work, we aim to provide a comprehensive analysis of LLMs' reasoning competence, specifically focusing on their prompt dependency. In particular, we introduce a new benchmark dataset with a series of simple reasoning questions demanding shallow logical reasoning. Aligned with cognitive psychology standards, the questions are confined to a basic domain revolving around geometric figures, ensuring that responses are independent of any pre-existing intuition about the world and rely solely on deduction. An empirical analysis involving zero-shot and few-shot prompting across 24 LLMs of different sizes reveals that, while LLMs with over 70 billion parameters perform better in the zero-shot setting, there is still a large room for improvement. An additional test with chain-of-thought prompting over 22 LLMs shows that this additional prompt can aid or damage the performance of models, depending on whether the rationale is required before or after the answer.

Federico Cabitza, Davide Ciucci, Gabriella Pasi et al.

This article discusses open problems, implemented solutions, and future research in the area of responsible AI in healthcare. In particular, we illustrate two main research themes related to the work of two laboratories within the Department of Informatics, Systems, and Communication at the University of Milano-Bicocca. The problems addressed concern, in particular, {uncertainty in medical data and machine advice}, and the problem of online health information disorder.

Elias Bassani, Marco Viviani

Nowadays, thanks to Web 2.0 technologies, people have the possibility to generate and spread contents on different social media in a very easy way. In this context, the evaluation of the quality of the information that is available online is becoming more and more a crucial issue. In fact, a constant flow of contents is generated every day by often unknown sources, which are not certified by traditional authoritative entities. This requires the development of appropriate methodologies that can evaluate in a systematic way these contents, based on `objective' aspects connected with them. This would help individuals, who nowadays tend to increasingly form their opinions based on what they read online and on social media, to come into contact with information that is actually useful and verified. Wikipedia is nowadays one of the biggest online resources on which users rely as a source of information. The amount of collaboratively generated content that is sent to the online encyclopedia every day can let to the possible creation of low-quality articles (and, consequently, misinformation) if not properly monitored and revised. For this reason, in this paper, the problem of automatically assessing the quality of Wikipedia articles is considered. In particular, the focus is on the analysis of hand-crafted features that can be employed by supervised machine learning techniques to perform the classification of Wikipedia articles on qualitative bases. With respect to prior literature, a wider set of characteristics connected to Wikipedia articles are taken into account and illustrated in detail. Evaluations are performed by considering a labeled dataset provided in a prior work, and different supervised machine learning algorithms, which produced encouraging results with respect to the considered features.