arXiv subjects

Antinisca Di Marco

Publications and source records attributed to Antinisca Di Marco.

At least 19 recordsLinked to original sources

arXiv2026

A Generalised Exponentiated Gradient Approach to Enhance Fairness in Binary and Multi-class Classification Tasks

The widespread use of AI and ML models in sensitive areas raises significant concerns about fairness. While the research community has introduced various methods for bias mitigation in binary classification tasks, the issue remains under-explored in multi-class classification settings. To address this limitation, in this paper, we first formulate the problem of fair learning in multi-class classification as a multi-objective problem between effectiveness (i.e., prediction correctness) and multiple linear fairness constraints. Next, we propose a Generalised Exponentiated Gradient (GEG) algorithm to solve this task. GEG is an in-processing algorithm that enhances fairness in binary and multi-class classification settings under multiple fairness definitions. We conduct an extensive empirical evaluation of GEG against six baselines across seven multi-class and three binary datasets, using four widely adopted effectiveness metrics and three fairness definitions.

cs.LG↗

arXiv2026

Smaller Models, Unexpected Costs: Trade-offs in LLM Quantization for Automated Program Repair

Large Language Models (LLMs) are powerful tools and have been increasingly adopted for complex software engineering tasks. As the number of parameters increases, results can often be improved, but this also imposes substantial memory requirements. While quantization effectively reduces the memory footprint, its overall impact is often summarized only by benchmark scores, which mask changes in model behavior and non-functional overheads. In this work, we conduct an empirical evaluation of LLM quantization using Automated Program Repair (APR), a complex task in software engineering. We analyze 13 quantization configurations spanning different bit-widths, methods, and target components (weights and KV-cache) across six representative LLMs, evaluated on two APR benchmarks (HumanEval-Java and Defects4J). Our findings reveal that base and quantized models can provide different sets of repaired problems with little overlap, while retaining a comparable number of repaired problems. Although quantization successfully reduces memory footprints by up to 85%, it increases both inference time and energy consumption, which we attribute to suboptimal hardware utilization. Our Pareto trade-off analysis shows that 48% of the configurations evaluated are strictly dominated by alternatives. Rather than identifying a superior quantization method, our findings highlight that the trade-offs between effectiveness, memory footprint, and energy efficiency are sensitive to the underlying model architecture and the complexity of the task.

cs.SE↗

arXiv2026

SafeTune: Search-based Harmfulness Minimisation for Large Language Models

The widespread adoption of Large Language Models (LLMs) raises concerns about the potential harmfulness of their responses. In this paper, we first investigate the harmfulness of responses from four general-purpose LLMs. Next, we propose SafeTune, a multi-objective search-based approach to mitigate harmfulness while increasing response relevance through hyperparameter tuning and system prompt engineering. Our initial evaluation shows that SafeTune significantly reduces the rate of harmful responses generated by Qwen3.5 0.8B and increases prompt-response relevance (both with a large effect size). Among the parameters we explore, we also find that encouraging greater repetition in responses is most impactful in reducing harmfulness while increasing relevance.

cs.SE↗

arXiv2025

REPAIR Approach for Social-based City Reconstruction Planning in case of natural disasters

Natural disasters always have several effects on human lives. It is challenging for governments to tackle these incidents and to rebuild the economic, social and physical infrastructures and facilities with the available resources (mainly budget and time). Governments always define plans and policies according to the law and political strategies that should maximise social benefits. The severity of damage and the vast resources needed to bring life back to normality make such reconstruction a challenge. This article is the extension of our previously published work by conducting comprehensive comparative analysis by integrating additional deep learning models plus random agent which is used as a baseline. Our prior research introduced a decision support system by using the Deep Reinforcement Learning technique for the planning of post-disaster city reconstruction, maximizing the social benefit of the reconstruction process, considering available resources, meeting the needs of the broad community stakeholders (like citizens' social benefits and politicians' priorities) and keeping in consideration city's structural constraints (like dependencies among roads and buildings). The proposed approach, named post disaster REbuilding plAn ProvIdeR (REPAIR) is generic. It can determine a set of alternative plans for local administrators who select the ideal one to implement, and it can be applied to areas of any extension. We show the application of REPAIR in a real use case, i.e., to the L'Aquila reconstruction process, damaged in 2009 by a major earthquake.

cs.CY↗

arXiv2025

How Do Generative Models Draw a Software Engineer? A Case Study on Stable Diffusion Bias

Generative models are nowadays widely used to generate graphical content used for multiple purposes, e.g. web, art, advertisement. However, it has been shown that the images generated by these models could reinforce societal biases already existing in specific contexts. In this paper, we focus on understanding if this is the case when one generates images related to various software engineering tasks. In fact, the Software Engineering (SE) community is not immune from gender and ethnicity disparities, which could be amplified by the use of these models. Hence, if used without consciousness, artificially generated images could reinforce these biases in the SE domain. Specifically, we perform an extensive empirical evaluation of the gender and ethnicity bias exposed by three versions of the Stable Diffusion (SD) model (a very popular open-source text-to-image model) - SD 2, SD XL, and SD 3 - towards SE tasks. We obtain 6,720 images by feeding each model with two sets of prompts describing different software-related tasks: one set includes the Software Engineer keyword, and one set does not include any specification of the person performing the task. Next, we evaluate the gender and ethnicity disparities in the generated images. Results show how all models are significantly biased towards male figures when representing software engineers. On the contrary, while SD 2 and SD XL are strongly biased towards White figures, SD 3 is slightly more biased towards Asian figures. Nevertheless, all models significantly under-represent Black and Arab figures, regardless of the prompt style used. The results of our analysis highlight severe concerns about adopting those models to generate content for SE tasks and open the field for future research on bias mitigation in this context.

cs.SE↗

arXiv2025

Investigating the Role of LLMs Hyperparameter Tuning and Prompt Engineering to Support Domain Modeling

The introduction of large language models (LLMs) has enhanced automation in software engineering tasks, including in Model Driven Engineering (MDE). However, using general-purpose LLMs for domain modeling has its limitations. One approach is to adopt fine-tuned models, but this requires significant computational resources and can lead to issues like catastrophic forgetting. This paper explores how hyperparameter tuning and prompt engineering can improve the accuracy of the Llama 3.1 model for generating domain models from textual descriptions. We use search-based methods to tune hyperparameters for a specific medical data model, resulting in a notable quality improvement over the baseline LLM. We then test the optimized hyperparameters across ten diverse application domains. While the solutions were not universally applicable, we demonstrate that combining hyperparameter tuning with prompt engineering can enhance results across nearly all examined domain models.

cs.SE↗

arXiv2025

Simplicity by Obfuscation: Evaluating LLM-Driven Code Transformation with Semantic Elasticity

Code obfuscation is the conversion of original source code into a functionally equivalent but less readable form, aiming to prevent reverse engineering and intellectual property theft. This is a challenging task since it is crucial to maintain functional correctness of the code while substantially disguising the input code. The recent development of large language models (LLMs) paves the way for practical applications in different domains, including software engineering. This work performs an empirical study on the ability of LLMs to obfuscate Python source code and introduces a metric (i.e., semantic elasticity) to measure the quality degree of obfuscated code. We experimented with 3 leading LLMs, i.e., Claude-3.5-Sonnet, Gemini-1.5, GPT-4-Turbo across 30 Python functions from diverse computational domains. Our findings reveal GPT-4-Turbo's remarkable effectiveness with few-shot prompting (81% pass rate versus 29% standard prompting), significantly outperforming both Gemini-1.5 (39%) and Claude-3.5-Sonnet (30%). Notably, we discovered a counter-intuitive "obfuscation by simplification" phenomenon where models consistently reduce rather than increase cyclomatic complexity. This study provides a methodological framework for evaluating AI-driven obfuscation while highlighting promising directions for leveraging LLMs in software security.

cs.SE↗

arXiv2025

An Evaluation Framework for the FAIR Assessment tools in Open Science

Open science represents a transformative research approach essential for enhancing sustainability and impact. Data generation encompasses various methods, from automated processes to human-driven inputs, creating a rich and diverse landscape. Embracing the FAIR principles -- making data and, in general, artifacts (such as code, configurations, documentation, etc) findable, accessible, interoperable, and reusable -- ensures research integrity, transparency, and reproducibility, and researchers enhance the efficiency and efficacy of their endeavors, driving scientific innovation and the advancement of knowledge. Open Science Platforms OSP (i.e., technologies that publish data in a way that they are findable, accessible, interoperable, and reusable) are based on open science guidelines and encourage accessibility, cooperation, and transparency in scientific research. Evaluating OSP will yield sufficient data and artifacts to enable better sharing and arrangement, stimulating more investigation and the development of new platforms. In this paper, we propose an evaluation framework that results from evaluating twenty-two FAIR-a tools assessing the FAIR principles of OSP to identify differences, shortages, and possible efficiency improvements.

cs.DB↗

arXiv2025

How fair are we? From conceptualization to automated assessment of fairness definitions

Fairness is a critical concept in ethics and social domains, but it is also a challenging property to engineer in software systems. With the increasing use of machine learning in software systems, researchers have been developing techniques to automatically assess the fairness of software systems. Nonetheless, a significant proportion of these techniques rely upon pre-established fairness definitions, metrics, and criteria, which may fail to encompass the wide-ranging needs and preferences of users and stakeholders. To overcome this limitation, we propose a novel approach, called MODNESS, that enables users to customize and define their fairness concepts using a dedicated modeling environment. Our approach guides the user through the definition of new fairness concepts also in emerging domains, and the specification and composition of metrics for its evaluation. Ultimately, MODNESS generates the source code to implement fair assessment based on these custom definitions. In addition, we elucidate the process we followed to collect and analyze relevant literature on fairness assessment in software engineering (SE). We compare MODNESS with the selected approaches and evaluate how they support the distinguishing features identified by our study. Our findings reveal that i) most of the current approaches do not support user-defined fairness concepts; ii) our approach can cover two additional application domains not addressed by currently available tools, i.e., mitigating bias in recommender systems for software engineering and Arduino software component recommendations; iii) MODNESS demonstrates the capability to overcome the limitations of the only two other Model-Driven Engineering-based approaches for fairness assessment.

cs.SE↗

arXiv2024

On the Compression of Language Models for Code: An Empirical Study on CodeBERT

Language models have proven successful across a wide range of software engineering tasks, but their significant computational costs often hinder their practical adoption. To address this challenge, researchers have begun applying various compression strategies to improve the efficiency of language models for code. These strategies aim to optimize inference latency and memory usage, though often at the cost of reduced model effectiveness. However, there is still a significant gap in understanding how these strategies influence the efficiency and effectiveness of language models for code. Here, we empirically investigate the impact of three well-known compression strategies -- knowledge distillation, quantization, and pruning -- across three different classes of software engineering tasks: vulnerability detection, code summarization, and code search. Our findings reveal that the impact of these strategies varies greatly depending on the task and the specific compression method employed. Practitioners and researchers can use these insights to make informed decisions when selecting the most appropriate compression strategy, balancing both efficiency and effectiveness based on their specific needs.

cs.SE↗

arXiv2024

Report on Female Participation in Informatics degrees in Europe

This study aims to enrich and leverage data from the Informatics Europe Higher Education (IEHE) data portal to extract and analyze trends in female participation in Informatics across Europe. The research examines the proportion of female students, first-year enrollments, and degrees awarded to women in the field. The issue of low female participation in Informatics has long been recognized as a persistent challenge and remains a critical area of scholarly inquiry. Furthermore, existing literature indicates that socio-economic factors can unpredictably influence female participation, complicating efforts to address the gender gap. The analysis focuses on participation data from research universities at various academic levels, including Bachelors, Masters, and PhD programs, and seeks to uncover potential correlations between female participation and geographical or economic zones. The dataset was first enriched by integrating additional information, such as each country's GDP and relevant geographical data, sourced from various online repositories. Subsequently, the data was cleaned to ensure consistency and eliminate incomplete time series. A final set of complete time series was selected for further analysis. We then used the data collected from the internet to assign countries to different clusters. Specifically, we employed Economic Zone, Geographical Area, and GDP quartile to cluster countries and compare their temporal trends both within and between clusters. We analyze the results for each classification and derive conclusions based on the available data.

cs.CY↗

arXiv2024

VAMP: Visual Analytics for Microservices Performance

Analysis of microservices' performance is a considerably challenging task due to the multifaceted nature of these systems. Each request to a microservices system might raise several Remote Procedure Calls (RPCs) to services deployed on different servers and/or containers. Existing distributed tracing tools leverage swimlane visualizations as the primary means to support performance analysis of microservices. These visualizations are particularly effective when it is needed to investigate individual end-to-end requests' performance behaviors. Still, they are substantially limited when more complex analyses are required, as when understanding the system-wide performance trends is needed. To overcome this limitation, we introduce vamp, an innovative visual analytics tool that enables, at once, the performance analysis of multiple end-to-end requests of a microservices system. Vamp was built around the idea that having a wide set of interactive visualizations facilitates the analyses of the recurrent characteristics of requests and their relation w.r.t. the end-to-end performance behavior. Through an evaluation of 33 datasets from an established open-source microservices system, we demonstrate how vamp aids in identifying RPC execution time deviations with significant impact on end-to-end performance. Additionally, we show that vamp can support in pinpointing meaningful structural patterns in end-to-end requests and their relationship with microservice performance behaviors.

cs.SE↗

arXiv2023

Towards a Prediction of Machine Learning Training Time to Support Continuous Learning Systems Development

The problem of predicting the training time of machine learning (ML) models has become extremely relevant in the scientific community. Being able to predict a priori the training time of an ML model would enable the automatic selection of the best model both in terms of energy efficiency and in terms of performance in the context of, for instance, MLOps architectures. In this paper, we present the work we are conducting towards this direction. In particular, we present an extensive empirical study of the Full Parameter Time Complexity (FPTC) approach by Zheng et al., which is, to the best of our knowledge, the only approach formalizing the training time of ML models as a function of both dataset's and model's parameters. We study the formulations proposed for the Logistic Regression and Random Forest classifiers, and we highlight the main strengths and weaknesses of the approach. Finally, we observe how, from the conducted study, the prediction of training time is strictly related to the context (i.e., the involved dataset) and how the FPTC approach is not generalizable.

cs.LG↗

arXiv2023

Data-Driven Analysis of Gender Fairness in the Software Engineering Academic Landscape

Gender bias in education gained considerable relevance in the literature over the years. However, while the problem of gender bias in education has been widely addressed from a student perspective, it is still not fully analysed from an academic point of view. In this work, we study the problem of gender bias in academic promotions (i.e., from Researcher to Associated Professor and from Associated to Full Professor) in the informatics (INF) and software engineering (SE) Italian communities. In particular, we first conduct a literature review to assess how the problem of gender bias in academia has been addressed so far. Next, we describe a process to collect and preprocess the INF and SE data needed to analyse gender bias in Italian academic promotions. Subsequently, we apply a formal bias metric to these data to assess the amount of bias and look at its variation over time. From the conducted analysis, we observe how the SE community presents a higher bias in promotions to Associate Professors and a smaller bias in promotions to Full Professors compared to the overall INF community.

cs.CY↗

arXiv2023

A Decision Tree to Shepherd Scientists through Data Retrievability

Reproducibility is a crucial aspect of scientific research that involves the ability to independently replicate experimental results by analysing the same data or repeating the same experiment. Over the years, many works have been proposed to make the results of the experiments actually reproducible. However, very few address the importance of data reproducibility, defined as the ability of independent researchers to retain the same dataset used as input for experimentation. Properly addressing the problem of data reproducibility is crucial because often just providing a link to the data is not enough to make the results reproducible. In fact, also proper metadata (e.g., preprocessing instruction) must be provided to make a dataset fully reproducible. In this work, our aim is to fill this gap by proposing a decision tree to sheperd researchers through the reproducibility of their datasets. In particular, this decision tree guides researchers through identifying if the dataset is actually reproducible and if additional metadata (i.e., additional resources needed to reproduce the data) must also be provided. This decision tree will be the foundation of a future application that will automate the data reproduction process by automatically providing the necessary metadata based on the particular context (e.g., data availability, data preprocessing, and so on). It is worth noting that, in this paper, we detail the steps to make a dataset retrievable, while we will detail other crucial aspects for reproducibility (e.g., dataset documentation) in future works.

cs.DL↗

arXiv2023

Architectural Support for Software Performance in Continuous Software Engineering: A Systematic Mapping Study

The continuous software engineering paradigm is gaining popularity in modern development practices, where the interleaving of design and runtime activities is induced by the continuous evolution of software systems. In this context, performance assessment is not easy, but recent studies have shown that architectural models evolving with the software can support this goal. In this paper, we present a mapping study aimed at classifying existing scientific contributions that deal with the architectural support for performance-targeted continuous software engineering. We have applied the systematic mapping methodology to an initial set of 215 potentially relevant papers and selected 66 primary studies that we have analyzed to characterize and classify the current state of research. This classification helps to focus on the main aspects that are being considered in this domain and, mostly, on the emerging findings and implications for future research

cs.SE↗

arXiv2022

Modeling Quality and Machine Learning Pipelines through Extended Feature Models

The recently increased complexity of Machine Learning (ML) methods, led to the necessity to lighten both the research and industry development processes. ML pipelines have become an essential tool for experts of many domains, data scientists and researchers, allowing them to easily put together several ML models to cover the full analytic process starting from raw datasets. Over the years, several solutions have been proposed to automate the building of ML pipelines, most of them focused on semantic aspects and characteristics of the input dataset. However, an approach taking into account the new quality concerns needed by ML systems (like fairness, interpretability, privacy, etc.) is still missing. In this paper, we first identify, from the literature, key quality attributes of ML systems. Further, we propose a new engineering approach for quality ML pipeline by properly extending the Feature Models meta-model. The presented approach allows to model ML pipelines, their quality requirements (on the whole pipeline and on single phases), and quality characteristics of algorithms used to implement each pipeline phase. Finally, we demonstrate the expressiveness of our model considering the classification problem.

cs.SE↗

arXiv2021

Architecture Design for Human-Driven Systems

This paper highlights humans' social and mobility behaviors' role in the continuous engineering of sustainable socio-technical systems. Our approach relates the humans' characteristics and intentions with the system's goals, and models such interaction. Such a modeling approach aligns the architectural design and associated quality of service (QoS) with humans' quality of experience (QoE). We design a simulation environment that combines agent-based social simulation (ABSS) with architectural models generated through a model-driven engineering approach. Our modeling approach facilitates choosing the best architectural model and system configuration to enhance both the humans' and system's sustainability. We apply our approach to the Uffizi Galleries crowd management system. Taking advantage of real data, we model different scenarios that impact QoE. We then assess various architectural models with different SW/HW configurations to propose the optimal model based on different scenarios concerning QoS-QoE requirements.

cs.SE↗