Alessio Orsino
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Alessio Orsino has a total of 10 co-authors. Here are the top co-authors by number of shared publications:
- Domenico Talia
University of Calabria
10 shared publications - Riccardo Cantini
Department of Informatics, Modeling, Electronics and Systems Engineering (DIMES), University of Calabria, 87036 Rende, Italy
10 shared publications - Fabrizio Marozzo
University of Calabria
7 shared publications - Paolo Trunfio
University of Calabria
6 shared publications - Loris Belcastro
University of Calabria
3 shared publications - Nicola Gabriele
University of Calabria
3 shared publications
Latest Publications
- Open Access
- 16 November 2026
- Published by Elsevier in Future Generation Computer Systems
A proxy-based interpretable model for energy estimation in edge AI via in-context learning
- 16 November 2026
- Published by Elsevier in Future Generation Computer Systems
Open AccessA proxy-based interpretable model for energy estimation in edge AI via in-context learning
- 2 July 2026
- Published by Springer Nature
- p. 19-31
Federated Learning in the Edge-Cloud Continuum: A Task-Based Approach with Colony
- 2 July 2026
- Published by Springer Nature
- p. 19-31
Federated Learning in the Edge-Cloud Continuum: A Task-Based Approach with Colony
- Top Cited PapersOpen Access
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 40622-40647
Navigating the Edge-Cloud Continuum: A State-of-Practice Survey
The edge-cloud continuum has emerged as a transformative paradigm that meets the growing demand for low-latency, scalable, end-to-end service delivery by integrating decentralized edge resources with centralized cloud infrastructures. Driven by the exponential growth of IoT-generated data and the need for real-time responsiveness, this continuum features multi-layered architectures. However, its adoption is hindered by infrastructural challenges, fragmented standards, and limited guidance for developers and researchers. Existing surveys rarely tackle practical implementation or recent industrial advances. This survey closes those gaps from a developer-oriented perspective, introducing a conceptual framework for navigating the edge-cloud continuum. We systematically examine architectural models, performance metrics, and paradigms for computation, communication, and deployment, together with enabling technologies and widely used edge-to-cloud platforms.We also discuss real-world applications in smart cities, healthcare, and Industry 4.0, as well as tools for testing and experimentation. Drawing on academic research and practices of leading cloud providers, this survey serves as a practical guide for developers and a structured reference for researchers, while identifying open challenges and emerging trends that will shape the future of the continuum.
- 12 March 2026
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Access
- Vol. 14, 40622-40647
Top Cited PapersOpen AccessNavigating the Edge-Cloud Continuum: A State-of-Practice Survey
The edge-cloud continuum has emerged as a transformative paradigm that meets the growing demand for low-latency, scalable, end-to-end service delivery by integrating decentralized edge resources with centralized cloud infrastructures. Driven by the exponential growth of IoT-generated data and the need for real-time responsiveness, this continuum features multi-layered architectures. However, its adoption is hindered by infrastructural challenges, fragmented standards, and limited guidance for developers and researchers. Existing surveys rarely tackle practical implementation or recent industrial advances. This survey closes those gaps from a developer-oriented perspective, introducing a conceptual framework for navigating the edge-cloud continuum. We systematically examine architectural models, performance metrics, and paradigms for computation, communication, and deployment, together with enabling technologies and widely used edge-to-cloud platforms.We also discuss real-world applications in smart cities, healthcare, and Industry 4.0, as well as tools for testing and experimentation. Drawing on academic research and practices of leading cloud providers, this survey serves as a practical guide for developers and a structured reference for researchers, while identifying open challenges and emerging trends that will shape the future of the continuum.
- Open Access
- 15 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 2294
Reproducibility Report for SC25 Paper MetoHash: A Memory-Efficient and Traffic-Optimized Hashing Index on Hybrid PMem-DRAM Memories
- 15 November 2025
- Published by Association for Computing Machinery (ACM)
- p. 2294
Open AccessReproducibility Report for SC25 Paper MetoHash: A Memory-Efficient and Traffic-Optimized Hashing Index on Hybrid PMem-DRAM Memories
- 5 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 13 (2) , 1989-2001
Dynamic Hashtag Recommendation in Social Media With Trend Shift Detection and Adaptation
Hashtag recommendation systems have emerged as a key tool for automatically suggesting relevant hashtags and enhancing content categorization and search. However, existing static models struggle to adapt to the highly dynamic nature of social media conversations, where new hashtags constantly emerge and existing ones undergo semantic shifts. To address these challenges, this article introduces hashtag recommendation by detecting and adapting to trend shifts (H-ADAPTS), a dynamic hashtag recommendation methodology that employs a trend-aware mechanism to detect shifts in hashtag usage—reflecting evolving trends and topics within social media conversations—and triggers efficient model adaptation based on a (small) set of recent posts. Additionally, the Apache storm framework is leveraged to support scalable and fault-tolerant analysis of high-velocity social data, enabling the timely detection of trend shifts. Experimental results from two real-world case studies, including the COVID-19 pandemic and the 2020 US presidential election, demonstrate the effectiveness of H-ADAPTS in providing timely and relevant hashtag recommendations by adapting to emerging trends, significantly outperforming existing solutions.
- 5 November 2025
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Computational Social Systems
- Vol. 13 (2) , 1989-2001
Dynamic Hashtag Recommendation in Social Media With Trend Shift Detection and Adaptation
Hashtag recommendation systems have emerged as a key tool for automatically suggesting relevant hashtags and enhancing content categorization and search. However, existing static models struggle to adapt to the highly dynamic nature of social media conversations, where new hashtags constantly emerge and existing ones undergo semantic shifts. To address these challenges, this article introduces hashtag recommendation by detecting and adapting to trend shifts (H-ADAPTS), a dynamic hashtag recommendation methodology that employs a trend-aware mechanism to detect shifts in hashtag usage—reflecting evolving trends and topics within social media conversations—and triggers efficient model adaptation based on a (small) set of recent posts. Additionally, the Apache storm framework is leveraged to support scalable and fault-tolerant analysis of high-velocity social data, enabling the timely detection of trend shifts. Experimental results from two real-world case studies, including the COVID-19 pandemic and the 2020 US presidential election, demonstrate the effectiveness of H-ADAPTS in providing timely and relevant hashtag recommendations by adapting to emerging trends, significantly outperforming existing solutions.
- Open Access
- 15 October 2025
- Published by Springer Nature in Machine Learning
- Vol. 114 (11) , 1-41
Benchmarking adversarial robustness to bias elicitation in large language models: scalable automated assessment with LLM-as-a-judge
The growing integration of Large Language Models (LLMs) into critical societal domains has raised concerns about embedded biases that can perpetuate stereotypes and undermine fairness. Such biases may stem from historical inequalities in training data, linguistic imbalances, or adversarial manipulation. Despite mitigation efforts, recent studies show that LLMs remain vulnerable to adversarial attacks that elicit biased outputs. This work proposes a scalable benchmarking framework to assess LLM robustness to adversarial bias elicitation. Our methodology involves: ( i ) systematically probing models across multiple tasks targeting diverse sociocultural biases, ( ii ) quantifying robustness through safety scores using an LLM-as-a-Judge approach, and ( iii ) employing jailbreak techniques to reveal safety vulnerabilities. To facilitate systematic benchmarking, we release a curated dataset of bias-related prompts, named CLEAR-Bias . Our analysis, identifying DeepSeek V3 as the most reliable judge LLM, reveals that bias resilience is uneven, with age, disability, and intersectional biases among the most prominent. Some small models outperform larger ones in safety, suggesting that training and architecture may matter more than scale. However, no model is fully robust to adversarial elicitation, with jailbreak attacks using low-resource languages or refusal suppression proving effective across model families. We also find that successive LLM generations exhibit slight safety gains, while models fine-tuned for the medical domain tend to be less safe than their general-purpose counterparts.
- 15 October 2025
- Published by Springer Nature in Machine Learning
- Vol. 114 (11) , 1-41
Open AccessBenchmarking adversarial robustness to bias elicitation in large language models: scalable automated assessment with LLM-as-a-judge
The growing integration of Large Language Models (LLMs) into critical societal domains has raised concerns about embedded biases that can perpetuate stereotypes and undermine fairness. Such biases may stem from historical inequalities in training data, linguistic imbalances, or adversarial manipulation. Despite mitigation efforts, recent studies show that LLMs remain vulnerable to adversarial attacks that elicit biased outputs. This work proposes a scalable benchmarking framework to assess LLM robustness to adversarial bias elicitation. Our methodology involves: ( i ) systematically probing models across multiple tasks targeting diverse sociocultural biases, ( ii ) quantifying robustness through safety scores using an LLM-as-a-Judge approach, and ( iii ) employing jailbreak techniques to reveal safety vulnerabilities. To facilitate systematic benchmarking, we release a curated dataset of bias-related prompts, named CLEAR-Bias . Our analysis, identifying DeepSeek V3 as the most reliable judge LLM, reveals that bias resilience is uneven, with age, disability, and intersectional biases among the most prominent. Some small models outperform larger ones in safety, suggesting that training and architecture may matter more than scale. However, no model is fully robust to adversarial elicitation, with jailbreak attacks using low-resource languages or refusal suppression proving effective across model families. We also find that successive LLM generations exhibit slight safety gains, while models fine-tuned for the medical domain tend to be less safe than their general-purpose counterparts.
- 22 September 2025
- Published by Springer Nature
- p. 433-445
A Parameter-Efficient Approach to Distilling Large Language Models via Meta-learning
- 22 September 2025
- Published by Springer Nature
- p. 433-445
A Parameter-Efficient Approach to Distilling Large Language Models via Meta-learning
- Open Access
Is Reasoning All You Need? Probing Bias in the Age of Reasoning Language Models
Reasoning Language Models (RLMs) have gained traction for their ability to perform complex, multi-step reasoning tasks through mechanisms such as Chain-of-Thought (CoT) prompting or fine-tuned reasoning traces. While these capabilities promise improved reliability, their impact on robustness to social biases remains unclear. In this work, we leverage the CLEAR-Bias benchmark, originally designed for Large Language Models (LLMs), to investigate the adversarial robustness of RLMs to bias elicitation. We systematically evaluate state-of-the-art RLMs across diverse sociocultural dimensions, using an LLM-as-a-judge approach for automated safety scoring and leveraging jailbreak techniques to assess the strength of built-in safety mechanisms. Our evaluation addresses three key questions: (i) how the introduction of reasoning capabilities affects model fairness and robustness; (ii) whether models fine-tuned for reasoning exhibit greater safety than those relying on CoT prompting at inference time; and (iii) how the success rate of jailbreak attacks targeting bias elicitation varies with the reasoning mechanisms employed. Our findings reveal a nuanced relationship between reasoning capabilities and bias safety. Surprisingly, models with explicit reasoning, whether via CoT prompting or fine-tuned reasoning traces, are generally more vulnerable to bias elicitation than base models without such mechanisms, suggesting reasoning may unintentionally open new pathways for stereotype reinforcement. Reasoning-enabled models appear somewhat safer than those relying on CoT prompting, which are particularly prone to contextual reframing attacks through storytelling prompts, fictional personas, or reward-shaped instructions. These results challenge the assumption that reasoning inherently improves robustness and underscore the need for more bias-aware approaches to reasoning design.
Open AccessIs Reasoning All You Need? Probing Bias in the Age of Reasoning Language Models
Reasoning Language Models (RLMs) have gained traction for their ability to perform complex, multi-step reasoning tasks through mechanisms such as Chain-of-Thought (CoT) prompting or fine-tuned reasoning traces. While these capabilities promise improved reliability, their impact on robustness to social biases remains unclear. In this work, we leverage the CLEAR-Bias benchmark, originally designed for Large Language Models (LLMs), to investigate the adversarial robustness of RLMs to bias elicitation. We systematically evaluate state-of-the-art RLMs across diverse sociocultural dimensions, using an LLM-as-a-judge approach for automated safety scoring and leveraging jailbreak techniques to assess the strength of built-in safety mechanisms. Our evaluation addresses three key questions: (i) how the introduction of reasoning capabilities affects model fairness and robustness; (ii) whether models fine-tuned for reasoning exhibit greater safety than those relying on CoT prompting at inference time; and (iii) how the success rate of jailbreak attacks targeting bias elicitation varies with the reasoning mechanisms employed. Our findings reveal a nuanced relationship between reasoning capabilities and bias safety. Surprisingly, models with explicit reasoning, whether via CoT prompting or fine-tuned reasoning traces, are generally more vulnerable to bias elicitation than base models without such mechanisms, suggesting reasoning may unintentionally open new pathways for stereotype reinforcement. Reasoning-enabled models appear somewhat safer than those relying on CoT prompting, which are particularly prone to contextual reframing attacks through storytelling prompts, fictional personas, or reward-shaped instructions. These results challenge the assumption that reasoning inherently improves robustness and underscore the need for more bias-aware approaches to reasoning design.
- 28 January 2025
- Published by Springer Nature
- p. 52-68
Are Large Language Models Really Bias-Free? Jailbreak Prompts for Assessing Adversarial Robustness to Bias Elicitation
- 28 January 2025
- Published by Springer Nature
- p. 52-68
Are Large Language Models Really Bias-Free? Jailbreak Prompts for Assessing Adversarial Robustness to Bias Elicitation
- Open Access
- 4 May 2024
- Published by Springer Nature in Journal of Big Data
- Vol. 11 (1) , 1-17
Xai-driven knowledge distillation of large language models for efficient deployment on low-resource devices
Large Language Models (LLMs) are characterized by their inherent memory inefficiency and compute-intensive nature, making them impractical to run on low-resource devices and hindering their applicability in edge AI contexts. To address this issue, Knowledge Distillation approaches have been adopted to transfer knowledge from a complex model, referred to as the teacher, to a more compact, computationally efficient one, known as the student. The aim is to retain the performance of the original model while substantially reducing computational requirements. However, traditional knowledge distillation methods may struggle to effectively transfer crucial explainable knowledge from an LLM teacher to the student, potentially leading to explanation inconsistencies and decreased performance. This paper presents DiXtill, a method based on a novel approach to distilling knowledge from LLMs into lightweight neural architectures. The main idea is to leverage local explanations provided by an eXplainable Artificial Intelligence (XAI) method to guide the cross-architecture distillation of a teacher LLM into a self-explainable student, specifically a bi-directional LSTM network.Experimental results show that our XAI-driven distillation method allows the teacher explanations to be effectively transferred to the student, resulting in better agreement compared to classical distillation methods,thus enhancing the student interpretability. Furthermore, it enables the student to achieve comparable performance to the teacher LLM while also delivering a significantly higher compression ratio and speedup compared to other techniques such as post-training quantization and pruning, which paves the way for more efficient and sustainable edge AI applications
- 4 May 2024
- Published by Springer Nature in Journal of Big Data
- Vol. 11 (1) , 1-17
Open AccessXai-driven knowledge distillation of large language models for efficient deployment on low-resource devices
Large Language Models (LLMs) are characterized by their inherent memory inefficiency and compute-intensive nature, making them impractical to run on low-resource devices and hindering their applicability in edge AI contexts. To address this issue, Knowledge Distillation approaches have been adopted to transfer knowledge from a complex model, referred to as the teacher, to a more compact, computationally efficient one, known as the student. The aim is to retain the performance of the original model while substantially reducing computational requirements. However, traditional knowledge distillation methods may struggle to effectively transfer crucial explainable knowledge from an LLM teacher to the student, potentially leading to explanation inconsistencies and decreased performance. This paper presents DiXtill, a method based on a novel approach to distilling knowledge from LLMs into lightweight neural architectures. The main idea is to leverage local explanations provided by an eXplainable Artificial Intelligence (XAI) method to guide the cross-architecture distillation of a teacher LLM into a self-explainable student, specifically a bi-directional LSTM network.Experimental results show that our XAI-driven distillation method allows the teacher explanations to be effectively transferred to the student, resulting in better agreement compared to classical distillation methods,thus enhancing the student interpretability. Furthermore, it enables the student to achieve comparable performance to the teacher LLM while also delivering a significantly higher compression ratio and speedup compared to other techniques such as post-training quantization and pruning, which paves the way for more efficient and sustainable edge AI applications