Navid Nobani
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Previous Organizations / Institutes
The organizations / institutes from the scholar's other articles in the past ten years
Navid Nobani has a total of 21 co-authors. Here are the top co-authors by number of shared publications:
- Fabio Mercorio
University of Milano-Bicocca
9 shared publications - Lorenzo Malandri
University of Milano-Bicocca
5 shared publications - Mario Mezzanzanica
University of Milano-Bicocca
4 shared publications - Andrea Seveso
University of Milano-Bicocca
2 shared publications - Erik Cambria
Nanyang technological university, Singapore, Singapore
2 shared publications - Filippo Pallucchini
Department of Statistics and Quantitative Methods University of Milan‐Bicocca Milan Italy
2 shared publications
Latest Publications
- 14 July 2026
- Published by Wiley in Expert Systems
- Vol. 43 (8)
Benchmarking Distributional Vector Similarity Measures: A Survey
Measuring semantic similarity between words or phrases is central to natural language processing, information retrieval and computational linguistics. Despite their importance, similarity and distance measures are typically chosen by default (e.g., cosine similarity) or in an ad hoc fashion, with little empirical justification. This lack of systematic evaluation creates two gaps: first, the absence of a comprehensive taxonomy of measures that spans set‐based, vector‐based and information‐theoretic approaches; second, the lack of task‐aware benchmarking that quantifies how these measures perform across different models and applications. In this paper, we address these gaps by comparing 15 similarity and distance measures on four NLP tasks (sentence similarity, kNN classification, correlation analysis and visualisation) using multiple benchmark datasets and embedding models. Our results reveal that the effectiveness of similarity measures varies substantially depending on the task and model, challenging the assumption that cosine similarity is universally optimal. These findings highlight the practical risk of relying on default measures and provide a principled basis for selecting similarity functions in NLP, information retrieval and related fields.
- 14 July 2026
- Published by Wiley in Expert Systems
- Vol. 43 (8)
Benchmarking Distributional Vector Similarity Measures: A Survey
Measuring semantic similarity between words or phrases is central to natural language processing, information retrieval and computational linguistics. Despite their importance, similarity and distance measures are typically chosen by default (e.g., cosine similarity) or in an ad hoc fashion, with little empirical justification. This lack of systematic evaluation creates two gaps: first, the absence of a comprehensive taxonomy of measures that spans set‐based, vector‐based and information‐theoretic approaches; second, the lack of task‐aware benchmarking that quantifies how these measures perform across different models and applications. In this paper, we address these gaps by comparing 15 similarity and distance measures on four NLP tasks (sentence similarity, kNN classification, correlation analysis and visualisation) using multiple benchmark datasets and embedding models. Our results reveal that the effectiveness of similarity measures varies substantially depending on the task and model, challenging the assumption that cosine similarity is universally optimal. These findings highlight the practical risk of relying on default measures and provide a principled basis for selecting similarity functions in NLP, information retrieval and related fields.
- Open Access
- 16 March 2026
- Published by Elsevier in Information Processing & Management
- Vol. 63 (6)
Synthetic data generation: A tertiary study
- 16 March 2026
- Published by Elsevier in Information Processing & Management
- Vol. 63 (6)
Open AccessSynthetic data generation: A tertiary study
- Top Cited Papers
- 24 February 2026
- Published by Elsevier in Information Fusion
- Vol. 133
Learning across modalities: a systematic survey of multimodal models for financial analysis
- 24 February 2026
- Published by Elsevier in Information Fusion
- Vol. 133
Top Cited PapersLearning across modalities: a systematic survey of multimodal models for financial analysis
- Open Access
- 1 December 2025
- Published by Oxford University Press (OUP) in The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences
- Vol. 80 (Supplement) , S222-S232
Participatory approach, dissemination, and implementation of research on aging: the Age-It experience
Objectives: As populations age, health, social, and long-term care policies must promote well-being, active participation, and inclusion in later life. This article examines how dissemination and participatory strategies can bridge research and practice, drawing on the Age-It Research Program—one of Europe’s largest transdisciplinary initiatives on aging. Methods: Age-It involves 1,000+ researchers across 10 “Spokes” and several cross-cutting Boards, engaging more than 100 stakeholder organizations. We describe the overarching dissemination and co-creation strategy developed by the Stakeholders and Dissemination Board, acknowledging the diverse participatory practices within Spokes. We term our approach incremental participatory, recognizing both early stakeholder involvement and the practical constraints of coordinating a program of this scale. Results: Our experience highlights three key lessons. First, establishing a shared language, timeframe, and platform is essential for transdisciplinary collaboration and stakeholder engagement. Second, dissemination and knowledge mobilization must be embedded throughout the research process rather than treated as a final stage activity, combining digital tools (e.g., WikiAge-It, the Age-It Dashboard) with offline, community-based events to foster inclusive participation. Third, while co-creation enhances legitimacy and relevance, sustaining active engagement requires significant investments in resources, infrastructures, and facilitation processes. Discussion: By integrating participatory research principles with implementation science frameworks, Age-It demonstrates how structured collaboration and knowledge mobilization can support evidence-based policies and practices. However, challenges remain, including uneven stakeholder engagement, time constraints, and the difficulty of translating complex scientific insights into actionable formats. We conclude that dissemination and participatory approaches in aging research must be seen as dynamic, iterative, and resource-intensive processes.
- 1 December 2025
- Published by Oxford University Press (OUP) in The Journals of Gerontology, Series B: Psychological Sciences and Social Sciences
- Vol. 80 (Supplement) , S222-S232
Open AccessParticipatory approach, dissemination, and implementation of research on aging: the Age-It experience
Objectives: As populations age, health, social, and long-term care policies must promote well-being, active participation, and inclusion in later life. This article examines how dissemination and participatory strategies can bridge research and practice, drawing on the Age-It Research Program—one of Europe’s largest transdisciplinary initiatives on aging. Methods: Age-It involves 1,000+ researchers across 10 “Spokes” and several cross-cutting Boards, engaging more than 100 stakeholder organizations. We describe the overarching dissemination and co-creation strategy developed by the Stakeholders and Dissemination Board, acknowledging the diverse participatory practices within Spokes. We term our approach incremental participatory, recognizing both early stakeholder involvement and the practical constraints of coordinating a program of this scale. Results: Our experience highlights three key lessons. First, establishing a shared language, timeframe, and platform is essential for transdisciplinary collaboration and stakeholder engagement. Second, dissemination and knowledge mobilization must be embedded throughout the research process rather than treated as a final stage activity, combining digital tools (e.g., WikiAge-It, the Age-It Dashboard) with offline, community-based events to foster inclusive participation. Third, while co-creation enhances legitimacy and relevance, sustaining active engagement requires significant investments in resources, infrastructures, and facilitation processes. Discussion: By integrating participatory research principles with implementation science frameworks, Age-It demonstrates how structured collaboration and knowledge mobilization can support evidence-based policies and practices. However, challenges remain, including uneven stakeholder engagement, time constraints, and the difficulty of translating complex scientific insights into actionable formats. We conclude that dissemination and participatory approaches in aging research must be seen as dynamic, iterative, and resource-intensive processes.
- 3 December 2024
- Published by Springer Nature in Cognitive Computation
- Vol. 17 (1) , 1-24
eXplainable AI for Word Embeddings: A Survey
- 3 December 2024
- Published by Springer Nature in Cognitive Computation
- Vol. 17 (1) , 1-24
eXplainable AI for Word Embeddings: A Survey
- Open Access
XAI meets LLMs: A Survey of the Relation between Explainable AI and Large Language Models
In this survey, we address the key challenges in Large Language Models (LLM) research, focusing on the importance of interpretability. Driven by increasing interest from AI and business sectors, we highlight the need for transparency in LLMs. We examine the dual paths in current LLM research and eXplainable Artificial Intelligence (XAI): enhancing performance through XAI and the emerging focus on model interpretability. Our paper advocates for a balanced approach that values interpretability equally with functional advancements. Recognizing the rapid development in LLM research, our survey includes both peer-reviewed and preprint (arXiv) papers, offering a comprehensive overview of XAI's role in LLM research. We conclude by urging the research community to advance both LLM and XAI fields together.
Open AccessXAI meets LLMs: A Survey of the Relation between Explainable AI and Large Language Models
In this survey, we address the key challenges in Large Language Models (LLM) research, focusing on the importance of interpretability. Driven by increasing interest from AI and business sectors, we highlight the need for transparency in LLMs. We examine the dual paths in current LLM research and eXplainable Artificial Intelligence (XAI): enhancing performance through XAI and the emerging focus on model interpretability. Our paper advocates for a balanced approach that values interpretability equally with functional advancements. Recognizing the rapid development in LLM research, our survey includes both peer-reviewed and preprint (arXiv) papers, offering a comprehensive overview of XAI's role in LLM research. We conclude by urging the research community to advance both LLM and XAI fields together.
- 7 November 2022
- Published by Springer Nature in Cognitive Computation
- Vol. 15 (2) , 613-644
ConvXAI: a System for Multimodal Interaction with Any Black-box Explainer
- 7 November 2022
- Published by Springer Nature in Cognitive Computation
- Vol. 15 (2) , 613-644
ConvXAI: a System for Multimodal Interaction with Any Black-box Explainer
- 25 October 2022
- Published by Elsevier in Information Processing & Management
- Vol. 60 (1)
A survey on XAI and natural language explanations
- 25 October 2022
- Published by Elsevier in Information Processing & Management
- Vol. 60 (1)
A survey on XAI and natural language explanations
- 1 May 2022
- Published by Elsevier in Information Fusion
- Vol. 81, 103-115
ContrXT: Generating contrastive explanations from any text classifier
- 1 May 2022
- Published by Elsevier in Information Fusion
- Vol. 81, 103-115
ContrXT: Generating contrastive explanations from any text classifier
- 8 January 2022
- Published by Springer Nature in Cognitive Computation
- Vol. 14 (2) , 749-763
Embeddings Evaluation Using a Novel Measure of Semantic Similarity
- 8 January 2022
- Published by Springer Nature in Cognitive Computation
- Vol. 14 (2) , 749-763
Embeddings Evaluation Using a Novel Measure of Semantic Similarity