Antonio Mastropaolo — PublicationsNew conference and journal papers across the AI × Software Engineering community.tag:antonio-mastropaolo.com,2026:publications-feed2026-05-04T21:25:57ZAntonio Mastropaoloamastropaolo@wm.edufeed.pytag:antonio-mastropaolo.com,2026:papers/afrin2026tokensNot All Tokens Matter: Data-Centric Optimization for Efficient Code Summarization2026-12-31T00:00:00Z2026-12-31T00:00:00ZSaima AfrinZaiyu ChengTushar SharmaAlexander SerebrenikMassimiliano Di PentaAntonio MastropaoloarXiv, 2026tag:antonio-mastropaolo.com,2026:papers/cheng2026systempromptsAn Empirical Study on the Effects of System Prompts in Instruction-Tuned Models for Code Generation2026-12-31T00:00:00Z2026-12-31T00:00:00ZZaiyu ChengAntonio MastropaoloarXiv, 2026tag:antonio-mastropaolo.com,2026:papers/haque2026multitaskParameter-Efficient Multi-Task Fine-Tuning in Code-Related Tasks2026-12-31T00:00:00Z2026-12-31T00:00:00ZMd Zahidul HaqueSaima AfrinAntonio MastropaoloarXiv, 2026tag:antonio-mastropaolo.com,2026:papers/mastropaolo2026editorialEditorial of the special issue in the journal of systems and software on reliable and secure large language models for software engineering2026-12-31T00:00:00Z2026-12-31T00:00:00ZAntonio MastropaoloPietro LiguoriGabriele BavotaJSS, 2026tag:antonio-mastropaolo.com,2026:papers/mastropaolo2026hallucinationThe Virtue of Hallucination: When AI Mistakes Make Software Safer2026-12-31T00:00:00Z2026-12-31T00:00:00ZAntonio MastropaoloIEEE Computer, 2026tag:antonio-mastropaolo.com,2026:papers/mastropaolo2026llmLLM-Powered Security Test Generation: Oracles, Vulnerability Probes, and Adversarial Inputs2026-12-31T00:00:00Z2026-12-31T00:00:00ZAntonio MastropaoloRick KuhnJeffrey M. VoasIEEE Computer, 2026tag:antonio-mastropaolo.com,2026:papers/mastropaolo2026mindMind the Overlap: Trustworthy Evaluation for Large Code Models2026-12-31T00:00:00Z2026-12-31T00:00:00ZAntonio MastropaoloIEEE Computer, 2026tag:antonio-mastropaolo.com,2026:papers/mastropaolo2026secretsSecrets in the Synapses: When Steganography Meets Large Language Models2026-12-31T00:00:00Z2026-12-31T00:00:00ZAntonio MastropaoloIEEE Computer, 2026tag:antonio-mastropaolo.com,2026:papers/pepe2026datasetsDatasets, bias, licenses, and terms of use: A large and longitudinal study on the documentation of hugging face machine learning models2026-12-31T00:00:00Z2026-12-31T00:00:00ZFederica PepeVittoria NardoneAntonio MastropaoloGerardo CanforaGabriele BavotaMassimiliano Di PentaEMSE, 2026tag:antonio-mastropaolo.com,2026:papers/rodriguez2026benchmarkingTowards Comprehensive Benchmarking Infrastructure for LLMs In Software Engineering2026-12-31T00:00:00Z2026-12-31T00:00:00ZDaniel Rodríguez-CárdenasXiaochang LiMarcos MacedoAntonio MastropaoloDipin KhatiYuan TianHuajie ShaoDenys PoshyvanykarXiv, 2026tag:antonio-mastropaolo.com,2026:papers/shahnami2026socToward Reliable Security Operations Center Testing With Foundation Models2026-12-31T00:00:00Z2026-12-31T00:00:00ZSophia ShahnamiAntonio MastropaoloJeffrey M. VoasIEEE Computer, 2026tag:antonio-mastropaolo.com,2026:papers/tufano2026devgenaiDevelopers and generative AI: A study of self-admitted usage in open source projects2026-12-31T00:00:00Z2026-12-31T00:00:00ZRosalia TufanoFederica PepeFiorella ZampettiAntonio MastropaoloOzren DabicMassimiliano Di PentaGabriele BavotaEMSE, 2026tag:antonio-mastropaolo.com,2025:papers/afrin2025peftA Systematic Literature Review of Parameter-Efficient Fine-Tuning for Large Code Models2025-12-31T00:00:00Z2025-12-31T00:00:00ZSaima AfrinMd Zahidul HaqueAntonio MastropaoloTOSEM, 2025tag:antonio-mastropaolo.com,2025:papers/afrin2025qloraResource-Efficient and Effective Code Summarization2025-12-31T00:00:00Z2025-12-31T00:00:00ZSaima AfrinJoseph CallKhai NguyenOscar ChaparroAntonio MastropaoloFORGE, 2025tag:antonio-mastropaolo.com,2025:papers/afrin2025quantizationIs Quantization a Deal-breaker? Empirical Insights from Large Code Models2025-12-31T00:00:00Z2025-12-31T00:00:00ZSaima AfrinBowen XuAntonio MastropaoloICSME, 2025tag:antonio-mastropaolo.com,2025:papers/casillo2025rationaleTowards Generating the Rationale for Code Changes2025-12-31T00:00:00Z2025-12-31T00:00:00ZFrancesco CasilloAntonio MastropaoloGabriele BavotaVincenzo DeufemiaCarmine GravinoICPC, 2025tag:antonio-mastropaolo.com,2025:papers/crupi2025llmjudgeOn the Effectiveness of LLM-as-a-Judge for Code Generation and Summarization2025-12-31T00:00:00Z2025-12-31T00:00:00ZGiuseppe CrupiRosalia TufanoAlejandro VelascoAntonio MastropaoloDenys PoshyvanykGabriele BavotaTSE, 2025tag:antonio-mastropaolo.com,2025:papers/fang2025smallerSmaller = Weaker? Benchmarking Robustness of Quantized LLMs in Code Generation2025-12-31T00:00:00Z2025-12-31T00:00:00ZSen FangWeiyuan DingAntonio MastropaoloBowen XuarXiv, 2025tag:antonio-mastropaolo.com,2025:papers/giagnorio2025quantizingQuantizing Large Language Models for Code Generation: A Differentiated Replication2025-12-31T00:00:00Z2025-12-31T00:00:00ZAlessandro GiagnorioAntonio MastropaoloSaima AfrinMassimiliano Di PentaGabriele BavotaarXiv, 2025tag:antonio-mastropaolo.com,2025:papers/mastropaolo2025cacheBreaking Bottlenecks in LLM Inference With Adaptive Cache Management2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio MastropaoloIEEE Computer, 2025tag:antonio-mastropaolo.com,2025:papers/mastropaolo2025databasesWhen Databases Age: How SQL Server and MySQL Handle the Test of Time2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio MastropaoloIEEE Computer, 2025tag:antonio-mastropaolo.com,2025:papers/mastropaolo2025heuristicsFrom Heuristics to Intelligence: Large Language Model-Driven Test Case Generation2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio MastropaoloRick KuhnJeffrey VoasIEEE Computer, 2025tag:antonio-mastropaolo.com,2025:papers/mastropaolo2025humanHuman or Machine? Rebuilding Trust in the Age of AI-Based Text Generation2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio MastropaoloIEEE Computer, 2025tag:antonio-mastropaolo.com,2025:papers/mastropaolo2025neurosymbolicA Path Less Traveled: Reimagining Software Engineering Automation via a Neurosymbolic Paradigm2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio MastropaoloDenys PoshyvanykAI-SDLC, 2025tag:antonio-mastropaolo.com,2025:papers/mastropaolo2025neurosymbolicmagCode, Chaos, and Clarity: Neurosymbolic Approaches to Trustworthy Software Automation2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio MastropaoloIEEE Computer, 2025tag:antonio-mastropaolo.com,2025:papers/mastropaolo2025pixelsPixels of Deception: How Evolutionary Algorithms Break AI Reliability2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio MastropaoloIEEE Computer, 2025tag:antonio-mastropaolo.com,2025:papers/mastropaolo2025promptPrompt Alchemy: Engineering the Magic of Code2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio MastropaoloIEEE Computer, 2025tag:antonio-mastropaolo.com,2025:papers/mastropaolo2025reshapingHow Artificial Intelligence Is Reshaping Our Lives2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio MastropaoloIEEE Computer, 2025tag:antonio-mastropaolo.com,2025:papers/mastropaolo2025selectiveSmarter, Not Harder: Efficient AI Training With Selective Data2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio MastropaoloIEEE Computer, 2025tag:antonio-mastropaolo.com,2025:papers/mastropaolo2025triumphFrom Triumph to Uncertainty: The Journey of Software Engineering in the AI Era2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio MastropaoloCamilo Escobar-VelasquezMario Linares-VasquezTOSEM, 2025tag:antonio-mastropaolo.com,2025:papers/mehditabar2025smartSmart but Costly? Benchmarking LLMs on Functional Accuracy and Energy Efficiency2025-12-31T00:00:00Z2025-12-31T00:00:00ZMohammad Javad MehditabarSaurabhsingh RajputAntonio MastropaoloTushar SharmaarXiv, 2025tag:antonio-mastropaolo.com,2025:papers/shu2025testloggingAn Empirical Study on Language Models for Generating Log Statements in Test Code2025-12-31T00:00:00Z2025-12-31T00:00:00ZHonglin ShuDong WangAntonio MastropaoloGabriele BavotaYasutaka KameiTOSEM, 2025tag:antonio-mastropaolo.com,2025:papers/velasco2025nspcToward Neurosymbolic Program Comprehension2025-12-31T00:00:00Z2025-12-31T00:00:00ZAlejandro VelascoAya GarryyevaDavid Nader PalacioAntonio MastropaoloDenys PoshyvanykICPC, 2025tag:antonio-mastropaolo.com,2025:papers/vitale2025coherenceOptimizing Datasets for Code Summarization: Is Code-Comment Coherence Enough?2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio VitaleAntonio MastropaoloRocco OlivetoMassimiliano Di PentaSimone ScalabrinoICPC, 2025tag:antonio-mastropaolo.com,2025:papers/vitale2025explainingToward Explaining Large Language Models in Software Engineering Tasks2025-12-31T00:00:00Z2025-12-31T00:00:00ZAntonio VitaleKhai-Nguyen NguyenDenys PoshyvanykRocco OlivetoSimone ScalabrinoAntonio MastropaoloarXiv, 2025tag:antonio-mastropaolo.com,2024:papers/mastropaolo2024ghwcomToward Automatically Completing GitHub Workflows2024-12-31T00:00:00Z2024-12-31T00:00:00ZAntonio MastropaoloFiorella ZampettiGabriele BavotaMassimiliano Di PentaICSE, 2024tag:antonio-mastropaolo.com,2024:papers/mastropaolo2024loggingLog Statements Generation via Deep Learning: Widening the Support Provided to Developers2024-12-31T00:00:00Z2024-12-31T00:00:00ZAntonio MastropaoloValentina FerrariLuca PascarellaGabriele BavotaJSS, 2024tag:antonio-mastropaolo.com,2024:papers/mastropaolo2024metricEvaluating Code Summarization Techniques: A New Metric and an Empirical Characterization2024-12-31T00:00:00Z2024-12-31T00:00:00ZAntonio MastropaoloMatteo CiniselliMassimiliano Di PentaGabriele BavotaICSE, 2024tag:antonio-mastropaolo.com,2024:papers/mastropaolo2024renamingAutomated Variable Renaming: Are We There Yet?2024-12-31T00:00:00Z2024-12-31T00:00:00ZAntonio MastropaoloEmad AghajaniLuca PascarellaGabriele BavotaEMSE, 2024tag:antonio-mastropaolo.com,2024:papers/mastropaolo2024risefallThe Rise and Fall (?) of Software Engineering2024-12-31T00:00:00Z2024-12-31T00:00:00ZAntonio MastropaoloCamilo Escobar-VelasquezMario Linares-VasquezSE2030, 2024tag:antonio-mastropaolo.com,2024:papers/mastropaolo2024snippetsTowards Summarizing Code Snippets Using Pre-Trained Transformers2024-12-31T00:00:00Z2024-12-31T00:00:00ZAntonio MastropaoloMatteo CiniselliLuca PascarellaRosalia TufanoEmad AghajaniGabriele BavotaICPC, 2024tag:antonio-mastropaolo.com,2024:papers/mastropaolo2024trainingHow the Training Procedure Impacts the Performance of Deep Learning-based Vulnerability Patching2024-12-31T00:00:00Z2024-12-31T00:00:00ZAntonio MastropaoloVittoria NardoneGabriele BavotaMassimiliano Di PentaEASE, 2024tag:antonio-mastropaolo.com,2024:papers/pepe2024huggingfaceHow do Hugging Face Models Document Datasets, Bias, and Licenses? An Empirical Study2024-12-31T00:00:00Z2024-12-31T00:00:00ZFederica PepeVittoria NardoneAntonio MastropaoloGerardo CanforaMassimiliano Di PentaGabriele BavotaICPC, 2024tag:antonio-mastropaolo.com,2024:papers/pepe2024satdA Taxonomy of Self-Admitted Technical Debt in Deep Learning Systems2024-12-31T00:00:00Z2024-12-31T00:00:00ZFederica PepeFiorella ZampettiAntonio MastropaoloGabriele BavotaMassimiliano Di PentaICSME, 2024tag:antonio-mastropaolo.com,2024:papers/scozzaro2024reformOn the Reform of the Italian Constitution: an Interdisciplinary Text Readability Analysis2024-12-31T00:00:00Z2024-12-31T00:00:00ZCalogero Jerik ScozzaroMatteo DelsantoAntonio MastropaoloEnrico MensaLuisa RevelliDaniele Paolo RadicioniNL4AI, 2024tag:antonio-mastropaolo.com,2024:papers/tufano2024chatgptUnveiling ChatGPT's Usage in Open Source Projects: A Mining-based Study2024-12-31T00:00:00Z2024-12-31T00:00:00ZRosalia TufanoAntonio MastropaoloFederica PepeOzren DabicMassimiliano Di PentaGabriele BavotaMSR, 2024tag:antonio-mastropaolo.com,2024:papers/tufano2024codereviewCode Review Automation: Strengths and Weaknesses of the State of the Art2024-12-31T00:00:00Z2024-12-31T00:00:00ZRosalia TufanoOzren DabicAntonio MastropaoloMatteo CiniselliGabriele BavotaTSE, 2024tag:antonio-mastropaolo.com,2023:papers/ciniselli2023codecompletionAn Empirical Study on the Usage of Transformer Models for Code Completion2023-12-31T00:00:00Z2023-12-31T00:00:00ZMatteo CiniselliNathan CooperLuca PascarellaAntonio MastropaoloEmad AghajaniDenys PoshyvanykMassimiliano Di PentaGabriele BavotaTSE, 2023tag:antonio-mastropaolo.com,2023:papers/mastropaolo2023copilotOn the Robustness of Code Generation Techniques: An Empirical Study on GitHub Copilot2023-12-31T00:00:00Z2023-12-31T00:00:00ZAntonio MastropaoloLuca PascarellaEmanuela GuglielmiMatteo CiniselliSimone ScalabrinoRocco OlivetoGabriele BavotaICSE, 2023tag:antonio-mastropaolo.com,2023:papers/mastropaolo2023satdTowards Automatically Addressing Self-Admitted Technical Debt: How Far Are We?2023-12-31T00:00:00Z2023-12-31T00:00:00ZAntonio MastropaoloMassimiliano Di PentaGabriele BavotaASE, 2023tag:antonio-mastropaolo.com,2023:papers/mastropaolo2023transferUsing Transfer Learning for Code-Related Tasks2023-12-31T00:00:00Z2023-12-31T00:00:00ZAntonio MastropaoloNathan CooperDavid Nader PalacioSimone ScalabrinoDenys PoshyvanykRocco OlivetoGabriele BavotaTSE, 2023tag:antonio-mastropaolo.com,2023:papers/rosa2023dockerfilesAutomatically Generating Dockerfiles via Deep Learning: Challenges and Promises2023-12-31T00:00:00Z2023-12-31T00:00:00ZGiovanni RosaAntonio MastropaoloSimone ScalabrinoGabriele BavotaRocco OlivetoICSSP, 2023tag:antonio-mastropaolo.com,2022:papers/mastropaolo2022usingUsing Deep Learning to Generate Complete Log Statements2022-12-31T00:00:00Z2022-12-31T00:00:00ZAntonio MastropaoloLuca PascarellaGabriele BavotaICSE, 2022tag:antonio-mastropaolo.com,2022:papers/tufano2022usingUsing Pre-Trained Models to Boost Code Review Automation2022-12-31T00:00:00Z2022-12-31T00:00:00ZRosalia TufanoSimone MasieroAntonio MastropaoloLuca PascarellaDenys PoshyvanykGabriele BavotaICSE, 2022tag:antonio-mastropaolo.com,2021:papers/mastropaolo2021commentAn Empirical Study on Code Comment Completion2021-12-31T00:00:00Z2021-12-31T00:00:00ZAntonio MastropaoloEmad AghajaniLuca PascarellaGabriele BavotaICSME, 2021tag:antonio-mastropaolo.com,2021:papers/mastropaolo2021t5Studying the Usage of Text-To-Text Transfer Transformer to Support Code-Related Tasks2021-12-31T00:00:00Z2021-12-31T00:00:00ZAntonio MastropaoloSimone ScalabrinoNathan CooperDavid Nader PalacioDenys PoshyvanykRocco OlivetoGabriele BavotaICSE, 2021tag:antonio-mastropaolo.com,2021:papers/scalabrino2021botAn Adaptive Search Budget Allocation Approach for Search-Based Test Case Generation2021-12-31T00:00:00Z2021-12-31T00:00:00ZSimone ScalabrinoAntonio MastropaoloRocco OlivetoGabriele BavotaTOSEM, 2021tag:antonio-mastropaolo.com,2013:papers/mastropaolo2013legalLegal documents categorization by compression2013-12-31T00:00:00Z2013-12-31T00:00:00ZAntonio MastropaoloFrancesco PallanteDaniele Paolo RadicioniICAIL, 2013