IIIA HUB

From November 10th to November 14th, the IIIA research group will be in Seoul for the 34th International Conference on Information and Knowledge Management (CIKM 2025). The team is excited to share their latest findings and engage with fellow researchers in the vibrant city of Seoul. Come join our presentations!

Monday, November 10th, 2025. Room 209B
During the PhD Symposium, Andrea Pasin will present an overview of his PhD research exploring how quantum computing could shape the future of information retrieval with the paper titled "Investigating the Usage and Evaluation of Quantum Computing Technologies for Information Access".
DOI: 10.1145/3746252.3761656

Tuesday, November 11th, 2025. 4:00 PM. Room 202
As the afternoon unfolds in Seoul, during the Full Paper session "LLM 3", Simone Merlo will present the paper titled "A Cost-Effective Framework to Evaluate LLM-Generated Relevance Judgements". This work, developed in collaboration with Stefano Marchesin, Guglielmo Faggioli, and Nicola Ferro, addresses how we can evaluate LLM-generated relevance judgments efficiently, saving on the cost of human annotators while maintaining solid statistical reliability.
DOI: 10.1145/3746252.3761200

Tuesday, November 11th, 2025. 5:45-7:30 PM. Room E, 3F (posterboard 99)
After Simone's presentation, join us at the poster session (posterboard 99), where Francesco L. De Faveri will present the paper titled "DP-COMET: A Differential Privacy Contextual Obfuscation MEchanism for Texts in Natural Language Processing". This work, developed in collaboration with Guglielmo Faggioli, and Nicola Ferro, explores how to protect user privacy using contextual obfuscation methods in NLP and IR applications.
GitHub: https://github.com/Kekkodf/DP-COMET
DOI: 10.1145/3746252.3760888

Thursday, November 13th, 2025. 4:00 PM. Room 203
In the last main conference afternoon, during the Full Paper session "Search & Retrieval 2" Ornella Irrera will present the paper titled "Scaling Trust: Veracity-Driven Defect Detection in Entity Search". This work, developed in collaboration with Stefano Marchesin, Gianmaria Silvello, and Omar Alonso (from Amazon Science), covers eRank, a new strategy for reranking in entity search, and ALADDIN, a scalable framework for defect detection in knowledge graphs.
GitHub: https://github.com/KGAccuracyEval/defect-detection4entity-search
DOI: 10.1145/3746252.3761208

This week, the IIIA research group has landed in Madrid for CLEF 2025, where the streets pulse with energy. The team has conducted the GutBrainIE (part of the BioASQ lab) and QuantumCLEF sessions.
Between sessions, the team has been inspired not just by talks, but by the rhythm of the city: from bustling cafés to colorful street art and hidden courtyards.
It's a reminder that research doesn't happen in isolation: ideas flourish when curiosity meets culture, and Madrid is providing plenty of both.

Several members of the IIIA research group are attending IIR 2025 in Cagliari this week.
Alongside presenting their latest research, the team has been exploring the city's historic streets and beautiful waterfront, enjoying the charm of Sardinia's capital.
It's been a week of science, collaboration, and soaking up the unique atmosphere of Cagliari.

Several members of the IIIA research group are attending the SEBD 2025 conference in Ischia this week. After presenting their research and taking part in interesting discussions, they also had the chance to visit the Aragonese Castle — a beautiful and historic place on the island. It has been a great opportunity to share ideas, meet other researchers, and enjoy the unique setting of the conference. A great mix of science, discussions, and just a little bit of summer mood.

On March 30, Ornella Irrera successfully defended her PhD thesis titled "Data search in practice: How to find scientific datasets and to link them to the literature". Ornella conducted her PhD project at the IIIA Hub under the supervision of prof. Gianmaria Silvello and Dr. Paolo Manghi from ISTI-CNR in Pisa.

Ornella will continue her research path as postdoc researcher at the IIIA Hub, working on dataset annotation and recommendation!

On February 27, prof. Gianmaria Silvello delivered a talk on Search Engines and Generative AI as part of the Advanced Training Course in Contemporary Archival Studies at the State Archives, within the module Public Archives and Digital Transformation. Nearly 50 archivists joined online, with many more attending in person in Rome, reflecting a strong interest in how emerging technologies can enhance access, preservation, and discovery in the archival field.

His take on the seminar: "Although I participated remotely this time, I am truly grateful for the invitation and the opportunity to discuss these exciting topics from a multidisciplinary perspective.".

From February 24 to February 27, Mirco Cazzaro, Ornella Irrera, and Laura Menotti participated at SWAT4HCLS 2025 (the 16th International Semantic Web Applications and tools for Health Care and Life Sciences Conference) in Barcelona, Spain. IIIA presented two papers: a Full Paper about the HEREDITARY Ontology for Genomic Data and and a Demo about MetaTron, an annotation tool for Biomedical Information Extraction. Both contributions recevied positive feedback and sparked some insightful and engaging discussions.

The full paper "HERO-Genomics: Bridging Genomic Data and Ontological Modelling" by Mirco Cazzaro (IIIA Hub), Ivo G. Gut (CNAG), Laura Menotti(IIIA Hub), Manuel Rueda (CNAG), and Gianmaria Silvello (IIIA Hub) described HERO-Genomics, the HEREDITARY ontology for Genomic Data. HERO-Genomics is a component of the broader Hereditary Ontology (HERO), developed to model the gut-brain connection from phenoclinical and genomic viewpoints. HERO serves as the backbone of the Semantic Data Integration platform for the HEREDITARY project, utilizing the Ontology-Based Data Access (OBDA) paradigm to query heterogeneous and distributed data respecting legal constraints. You can find the preprint here: preprint.

The demo "MetaTron: Streamlining Collaborative Annotation for Biomedical Documents" by Ornella Irrera, Stefano MArchesin, and Gianmaria Silvello (all from the IIIA Hub) describes MetaTron, a web-based collaborative tool that supports mention-level and document-level annotations and automatic built-in predictions. You can find the preprint here: preprint.

From February 20 to February 21, prof. Giorgio Maria Di Nunzio, prof. Gianmaria Silvello, and Laura Menotti participated at IRCDL 2025 (the 21st Conference on Information and Research Science Connecting to Digital and Library Science) in Udine, Italy. IIIA presented two papers: one in the Computer Science Track and one in the Digital Humanities Track in collaboration with the University of Bari (UniBa). Both contributions recevied positive feedback and won some prizes!

The paper "Extending Nanopublications with Knowledge Provenance for Multi-Source Scientific Assertions" by Fabio Giachelle, Stefano Marchesin, Laura Menotti, and Gianmaria Silvello (all from the IIIA Hub) won the Best Paper Award for the Computer Science Track. You can find the preprint here: preprint.

The paper "Knowledge Graphs for the Web Economy: The CHiPS&BITS Project on Cultural Heritage" by Stefano Ferilli (UniBa), Eleonora Bernasconi (UniBa), Domenico Redavid (UniBa), and Giorgio Maria Di Nunzio (IIIA Hub) won the Honorable Mention Award for the Digital Humanities Track.

From February 9 to February 12, prof. Nicola Ferro participated at SWIRL (the 4th Strategic Workshop on Information Retrieval in Lorne), an invite-only workshop gathering about 60 researchers from all over the world to envision the future of IR in the coming years. This year SWIRL was located at RMIT University in Melbourne, Australia. The agenda featured three days of intense brainstorming and discussions of the next 5 years research for IR and surroundings.

After SWIRL, a series of events were held with gathering at RMIT with further talks, discussions, and a poster session on cutting edge research topics. Laura Dietz focused on text-based machine learning and information retrieval as well as data science on watersheds. Nick Craswell presented the TREC 2024 RAG track and how to evaluate RAG systems using nuggets.Mohammad Aliannejadi talked about conversational search systems while exploring various areas in IR, machine learning, and NLP.

On January 21, prof. Nicola Ferro presented a session titled "Rappresentazione dei dati e della conoscenza e AI generativa" (Data and Knowledge Representation and Generative AI) during the training course "Introduzione all'intelligenza artificiale in medicina" (Introduction to Artificial Intelligence in Medicine) held at the University Hospital in Padova. The event was organized by the Department of Medicine of the University of Padova (DIMED) for the continuous education in healthcare professions.

The training course aims to explore the state of the art of AI in medicine, fostering dialogue among researchers, clinicians, and engineers to discuss the practical applications of AI in various specialist fields and analyze the benefits and risks associated with its use in clinical practice. This event is an opportunity to outline future perspectives in this rapidly evolving field, emphasizing how AI can be used to improve the quality of care while ensuring basic literacy for those approaching this sector. More information about the training course can be found in the Course Program (in italian).

The UK electronic information Group (UKeiG) is pleased to announce that the winner of the prestigious international Tony Kent Strix Memorial Award for 2024 is one of the world’s leading scientists in information retrieval, Nicola Ferro, Professor in Computer Science at the Department of Information Engineering of University of Padua. The award celebrates his distinguished research record and sustained, cumulative contribution to the field of information retrieval (IR).

The Strix award honours outstanding contributions to the field in memory of Tony Kent, a remarkable IR pioneer. It was inaugurated in 1998 by the Institute of Information Scientists and is now presented by UKeiG in partnership with the International Society for Knowledge Organisation UK (ISKO UK), the Royal Society of Chemistry Chemical Information and Computer Applications Group (RSC CICAG) and the British Computer Society Information Retrieval Specialist Group (BCS IRSG). Whereas at the time of Kent’s ground-breaking achievements IR was a fascinating sideline for relatively few specialists, nowadays the Search function underpins everyday living at home, at work and across the globe. This Award is more widely relevant than ever before.

Nicola’s IR research focuses on evaluation, from both a theoretical and an experimental perspective. He chairs the Steering Committee of CLEF (Conference and Labs of the Evaluation Forum), the internationally renowned European initiative for the evaluation of multilingual and multimodal information access systems, and through his efforts has made it central to the IR evaluation community. He was inducted into the Special Interest Group in Information Retrieval (SIGIR) Academy in 2023 in recognition of his significant research, innovation and service delivery.

The Strix judging panel noted that his contributions to IR are extensive, evidenced by an outstanding and impactful publications record. It would like to congratulate him on his prolific and significant leadership and contribution to the profession on multiple fronts.

Nicola was delighted to receive the news and will celebrate his award in a special free Zoom lecture on the afternoon of Thursday January 9th 2025, where he will reflect on the challenges and opportunities of generative AI for the IR community.

Book here for your free place:
https://www.cilip.org.uk/events/EventDetails.aspx?id=1913827&group=201314

"I’m honoured to receive this prestigious award and thank the Strix judging panel for such important recognition. This achievement would have not been possible without the mentors I've encountered throughout my career and without the collaboration with and inspiration from numerous colleagues, in Padua and around the world. I consider this award an appreciation of not only my work but also of the contributions of the international CLEF community."

Professor Norbert Fuhr, Information Engineering, Department of Computer Science and Applied Cognitive Science at the University of Duisburg-Essen in Germany, submitted the nomination. It included five letters of support from well-known scientists, one of whom was 2017 Strix award winner Professor Maarten de Rijke, University of Amsterdam Informatics.

"Early in his career, Nicola Ferro made ground-breaking contributions on reference models for digital libraries, on annotations of digital content, and on system design for digital libraries. His work in this area has been very influential. He is best known for his impressive contributions to the evaluation of information retrieval systems. He is among the world’s top experts in the field. Thanks to Nicola’s tireless work, his people skills, and his scientific vision, CLEF unites a broad set of tasks, from traditional document retrieval to environmental image classification to medical data integration. CLEF has been, and continues to be, instrumental in supporting and advancing the research agendas of countless researchers."

Linkedin post here: linkedin.com/posts/iiia-unipd_nicola-ferro-2024-tony-kent-memorial-award

On December 7th, prof. Stefano Marchesin held a workshop for the Global Citizenship Programme of the University of Padua titled "Generative AI: Navigating the Future of Content Creation in Society". The workshop aimed to introduce participants to the rapidly evolving world of generative AI and explore its implications in content creation. By the end of the module, participants had a better understanding of how AI is impacting on the media landscape, the potential benefits, but also the associated risks -- such as misinformation, hoaxes, and deepfakes.

The workshop explored generative AI, discussing its benefits -- such as supporting creativity and productivity across fields like journalism, art, and media -- but also its risks and challenges -- including a decrease in content quality and a growing threat of misinformation, hoaxes, and fake news. The pros and cons of AI content creation were explored with participants by stimulating discussions and experiences coming from different backgrounds. Special attention was given to the importance of critically evaluating media contents so as to counteract misinformation.

The Alumni association of the University of Padua is pleased to announce that Andrea Pasin, one of our PhD Students, won the Graduates Alumni Award for the Academic Year 2022/2023. The award celebrates the top master's graduates from the University of Padua's eight schools, based on merit and timely completion of the degree.

Andrea Pasin completed his master degree in Computer Engineering in just one year and a half and with the highest score with a thesis conducted at the IIIA Hub titled "QuantumCLEF: A Shared-Task Proposal to Evaluate the Performance of Quantum Computing for Information Retrieval Systems", under the supervision of prof. Nicola Ferro. He is now part of our team as a second-year PhD student supervised by prof. Nicola Ferro, working on Quantum Computing approaches applied to the Information Retrieval field.

On December 3, prof. Giorgio Maria Di Nunzio and prof. Federica Vezzani presented a session titled "Termsplaining for AI in Technical Communication" during the AI Open Forum held at Zanothel Europa, Bologna.

Through concrete examples and academic studies, they explored how accurate terminology can enhance AI's ability to understand technical texts, translate specialized documents, and support experts in a wide range of activities.

AI Open Forum is the first conference dedicated to Artificial Intelligence designed to address and focus on the most relevant aspects of Communication, Technical Documentation, and Intelligent Information, with continuous references to the competencies, skills, tasks, and responsibilities of Technical Communicators. More information can be found at the AI Open Forum Website (in italian).

From November 28 to November 30, prof. Nicola Ferro and prof. Gianmaria Silvello were invited for a 3-day retreat in the beautiful San Servolo Centro Venezia Eventi e Congressi in Venice to brainstorm on the next "Piano Regionale Prevenzione" (Regional Prevention Plan) by the Regione Veneto. The “Officina della Prevenzione” (roughly Prevention Workshop) brought together lots of different competences in the healthcare sector and beyond.

The first day consisted on an introductory plenary sessions and panels. In the second day, the audience was split into work groups and prof. Nicola Ferro participated to the “Le emergenze che impattano sulla salute pubblica: strumenti di simulazione ed esercitazione per rafforzare la preparedness” (Simulation exercises to enhance preparedness and readiness during emergencies impacting public health) where he tried to contribute with an ICT and information access point of view. On Saturday, a final plenary session distilled the output of the work groups.

It was a very stimulating experience and an opportunity to look at computer science and IR competencies from a very different perspective. Over the three days the discussion was about: intersectorial networks, diagnostic therapeutic assistance pathways and prevention, health and environment, integrated security, simulation exercises to increase preparedness in emergency management, one-health and antimicrobial resistance, information systems, and governance.

Lots of discussions and ideas focused on how information access methodologies can provide substantial innovation in these transversal areas, ranging from various forms of data integration and information extraction approaches, generative AI to improve simulation exercises, generative AI as a co-pilot during emergency management, various form of LLM-based monitoring, analysis, and personalization in communication with general public, and much more.

In November, the IIIA research group had the privilege of hosting Dr. Omar Alonso from Amazon Science (Palo Alto, CA, USA), renowned for his contributions to data science and knowledge discovery. His two-week visit was nothing short of inspiring, marked by fruitful collaborations and valuable knowledge-sharing sessions. Dr. Alonso’s visit has left a lasting impact on IIIA, fostering new research avenues and enhancing the academic growth of our community. We are grateful for his contributions and look forward to potential future collaborations that will continue to push the boundaries of scientific discovery.

In addition to his hands-on research efforts, Dr. Alonso also enriched the academic environment at department of Information Engineering at the Università degli Studi di Padova by delivering a guest lecture titled "Search in the LLMs Era".
The room was packed with enthusiastic students from the Graph Databases course (Master's in Computer Engineering) and Fundamentals of Information Systems (Master's in Data Science). This event offered a unique opportunity to dive into cutting-edge research shaping the future of search technologies in the era of large language models.

Key Highlights:

• Insights into groundbreaking research at Amazon Science
• Internship opportunities for ambitious students
• An engaging Q&A session where Omar Alonso shared his expertise

This seminar exemplifies the University of Padua’s unwavering commitment to providing exceptional learning experiences and fostering strong connections with leading tech innovators. By bridging academia and industry, we empower our students to excel in the global tech landscape.

To Dr. Omar Alonso: thank you for sharing your time, expertise, and passion for research with us!

This November, the IIIA research group had the immense honor of hosting Prof. Kalervo Järvelin from Tampere University (Finland), a luminary in the field of Information Retrieval (IR) evaluation. With a career dedicated to advancing IR methodologies and a legacy of impactful research, Prof. Järvelin’s visit was a remarkable opportunity for collaboration and learning. Prof. Järvelin is celebrated worldwide for his groundbreaking contributions, particularly as one of the authors of the nDCG (normalized Discounted Cumulative Gain), a fundamental metric for evaluating search engine effectiveness. His extensive work in IR evaluation has set benchmarks for the field, and in 2018, his pioneering contributions were recognized with the prestigious ACM SIGIR Gerard Salton Award, one of the highest honors in the domain of information retrieval.

During his time at IIIA, Prof. Järvelin shared his unparalleled expertise and engaged in thought-provoking discussions with our researchers. His visit not only fostered new ideas for ongoing projects but also underscored the importance of rigorous evaluation in information retrieval systems.

A highlight of his stay was his seminar titled "A Framework for IR Evaluation Integrating Task and User Characteristics", where he presented his latest work on how to include multi-theme relevance and user aspects within the IR evaluation framework.

Abstract of the talk:
Traditional search result evaluation metrics in information retrieval, such as MAP and NDCG, naively focus on topical relevance between a document and search topic and assume this relationship as mono-dimensional and often binary. They neglect document content overlap and assume gains piling up as the searcher examines the ranked list at greater length. We propose a novel search result evaluation framework based on multidimensional, graded relevance assessments, explicit modelling of document overlaps and attributes affecting document usability beyond relevance in task performance. Document relevance to a search task is seen to consist of several content themes and document usability attributes. Documents may also overlap regarding their content themes. Attributes such as document readability, trustworthiness, or language represent the entire document’s usability in the search task context, for a given searcher and their motivating task. The proposed framework evaluates the quality of a ranked search result, taking into account the contribution of each successive document, with estimated overlap across themes, and usability based on its attributes.

Reference:
Järvelin, K., and Sormunen, E. 2024. A Blueprint of IR Evaluation Integrating Task and User Characteristics: Proposal for Evaluation Framework. ACM Transactions on Information Systems, Volume 42, Issue 6, Article No.: 164, Pages 1-38; DOI: doi.org/10.1145/3675162.