Accepted papers

Regular papers

• Capisci a me: The Hidden Risks of Regional Language Processing in LLMs
  Giuseppe Magazzù, Giulia Rizzi and Elisabetta Fersini
  
• Don’t Say It!: Constraints, Compliance, and Communication when Language Models Play Taboo
  Sara Candussio, Francesca Padovani, Daniel Scalena and Malvina Nissim
  
• Più di Mille Parole? Multimodal and Multilingual Identification of Misogyny Cues in Internet Memes: A Case Study of English, Chinese, and Italian Datasets
  Minggui Duan and Alberto Barrón-Cedeño
  
• A Comparative Analysis of Datasets for Privacy Risk Detection in Online Content
  Gaia Gambarelli, Aldo Gangemi, Giovanni Livraga, Biagio Spiezia, Rocco Tripodi and Marco Viviani
  
• A Factorial Study of Synthetic Data Generation for Low-Resource Machine Translation using Grammar Books
  Varun Ghat Ravikumar, Sina Ahmadi, Lena Jäger and Rico Sennrich
  
• A Logical/Bayesian Backbone for Explainable Argumentation Dynamics in Conversational AI
  Raffaele Di Maso, Emilia Tanda, Domenico Morra, Maria Di Maro and Antonio Origlia
  
• Adapting Grounded Semantic Role Labelling to Italian Situated Robot Commands
  Federico Rosati, Claudiu Daniel Hromei, Antonio Scaiella, Danilo Croce and Roberto Basili
  
• Advocacy Radar: Monitoring Advocacy for Digital Rights
  Giulio Corradi and Marta Marchiori Manerba
  
• AlBERTurin: A Fully Open Family of Italian Encoder Models with Modern Architectures
  Matteo Rinaldi, Marco Madeddu, Calogero Jerik Scozzaro, Matteo Delsanto, Daniele P. Radicioni and Viviana Patti
  
• Before the Storm: Estimating Offensive Speech Risk from Italian News on YouTube
  Samuele D’Avenia, Eliana Di Palma, Lia Draetta, Soda Marem Lo and Marta Marchiori Manerba
  
• Beyond Concreteness: the Role of Semantic Specificity in Predicting Lexical Complexity in Italian
  Laura Occhipinti and Marianna Marcella Bolognesi
  
• Beyond Language Specialization: A RAG Evaluation of Italian LLMs
  Riccardo Agabiti, Giovanni Trappolini, Salvatore Rago, Raniero Romagnoli, Cristina Giannone and Fabrizio Silvestri
  
• Brownie IS-A cake, but sugar is not: Can encoders annotate and clean children’s (mis)built semantic hierarchies?
  Giovanni Puccetti, Adele Loia and Andrea Esuli
  
• Can LLMs Match Folk Wisdom? A Contextual Evaluation of Proverb Selection in Romance Dialects of Italy
  Gabriele Errico, Alan Ramponi, Sara Tonelli and Elisabetta Jezek
  
• Can Politics be Schizophrenic? Metaphorical and Stigmatizing Use of Psychiatric Terms in Italian News Media
  Ginevra Martinelli, Chiara Barattieri di San Pietro, Marta Bosia and Valentina Bambini
  
• CHALLENGE-IT: A Multi-Source Italian Benchmark with Hard Negatives for LLM-Generated Text Detection
  Alessandro Pisent and Fabrizio Silvestri
  
• Coherence, Cohesion and Complexity (Co3): a hybrid suite for text analysis
  Alessandro Maisto
  
• Coming and Going: Assessing Italian Motion Verb Usage in Language Models
  Maria Pia di Buono, Raffaele Manna, Silvio Calderaro, Mariapia Battipaglia and Johanna Monti
  
• Comparing Large Language Models and Humans in Historical Interpretation
  Fabio Celli and Georgios Spathoulas
  
• Computational Morphology for Norwegian Medical Compounds: An HMM, Neural, and Neurosymbolic Approach to Morpheme Boundary Detection
  Keziah Reina
  
• CorefUD Goes Latin: Enriching a Large Multilingual Coreference Resource with a Classical Language
  Roberta Leotta, Marco Passarotti, Martin Popel and Zdenek Zabokrtsky
  
• CoSITA: a Benchmark for Temporally Grounded State-Transition Understanding
  Andrea Pedrotti and Andrea Esuli
  
• Creating PFB: Prometeia Financial Benchmark
  Alessandro Pietro Bardelli, Tolga Cekic, Irem Demirtas, Marilena Di Bari, Michele Filannino and Simona Scala
  
• DA-Self-AMPLIFY: Improving In-Context Learning with Post-hoc Explanations for Subjective NLP
  Benedetta Muscato, Gizem Gezici and Fosca Giannotti
  
• Detecting Semantic Variation between German Standard Varieties through Word Embedding Comparison
  luca ducceschi and Federico Villa
  
• Developing a Perspectivist Historical Databank with Large Language Models
  Fabio Celli
  
• Dialogue moves and user engagement in Human-LLM Conversational Exchanges
  Martina Di Bratto, Valentina Russo, Azzurra Mancini, Marco Grazioso and Valeria Mauro
  
• Discrete Diffusion Language Models: An Experimental Analysis
  Thomas Bertolani, Davide Bucciarelli, Leonardo Zini, Marcella Cornia and Lorenzo Baraldi
  
• Do Language Models Know Their Slang? Queer Slang Understanding in User-Generated Content
  Arianna Denitto and Beatrice Savoldi
  
• Do LLMs Know Basic Italian? Evaluating Lexical Competence through Dictionary Entry Generation against the De Mauro Basic Vocabulary
  Manuel Gozzi, Guido Vetere and Francesca Fallucchi
  
• Do LLMs Truly Reason About Causes? Introducing AITIA, a Contrastive Benchmark for Commonsense Causal Reasoning
  Serena Auriemma, Martina Miliani and Alessandro Lenci
  
• Do LVLMs See Stereotypes as We Do? A Social Science Framework for Evaluating LVLMs’ Recognition and Explanation of Stereotypes in Images
  Marco Madeddu, Chiara Ferrando, Elisa Baldazzi, Horacio Jarquin-Vasquez, Marina Della Giusta and Viviana Patti
  
• Do We Speak the Same Language? Language, Language Models, and What We’re Talking About
  Malvina Nissim
  
• Dromedario-3: Localizing Tülu 3 Dataset to Italian
  Luca Gioffré, Marina Iuliana Aur, Francesco Ortame, Luca Moroni, Alberte Fernández-Castro, Elena Marafatto and Roberto Navigli
  
• EDEN: A Large-Scale Corpus of Clinical Notes for Italian
  Tiziano Labruna, Guido Bertolini, Pietro Ferrazzi and Bernardo Magnini
  
• Empathy Editing for Italian Medical Question Answering: Can Empathy Be Added Safely?
  Ahmad Reza Yoozbashi Zadeh, Daniela Occhipinti and Alessio Palmero Aprosio
  
• Enhanced Retrieval-Augmented Generation for Italian Public Administration Document Drafting
  Giovanni Valer and Alessio Palmero Aprosio
  
• Evaluating ASR Architectures for South Tyrolean Dialect to Standard German Translation
  luca ducceschi and Greta H. Franzini
  
• Evaluating Italian Readability-Controlled Text Simplification: A Multi-Dimensional Evaluation Framework
  Gaia Federica Francesca Ferrara, Michele Papucci, Felice Dell’Orletta and Giulia Venturi
  
• Evaluating LLMs on Target Sentiment Analysis and Emotion Recognition in German and Italian Journalism
  Gloria Comandini, Carlo Strapparava, Federico Collaoni, Charles Biagioni, Sabrina Turco Liveri and Roberto Zanoli
  
• Evaluating Transformer Representations of Italian Modal Constructions with Collostructional Analysis and Sentence Embeddings
  Giulia Calvi and Francesco Mambrini
  
• Exploring CompL-it: Quantitative and Graph-based Profiling of an Open Italian Lexical-Semantic Resource
  Emiliano Giovannetti, Andrea Bellandi, Simone Marchi and Mafalda Papini
  
• FAIR-ISLE: A Bilingual Platform for AI-mediated Inclusive Rewriting and Critical Language Awareness
  Francesca Filograsso, Lucia Siciliani, Rosita Belinda Maglie, Pierpaolo Basile and Marco DeGemmis
  
• Fine-tuning Large Language Models on Justinian’s Digest for Latin-to-Italian Specialised Translation
  Eleonora Ghizzota, Pierpaolo Basile, Lucia Siciliani, Paola Marongiu, Alessandra Cinini and Eva Sassolini
  
• From Equivalence to Label Flips: A Study of Semantic Consistency in Language Models Using Controlled Linguistic Transformations
  Guillermo Villate, Marco W. Shappeck, Borja Sanz Urquijo and Javier Del Ser
  
• From Prompt Engineering to Stimulus Validity: Semantic Constraint Evaluation in Text-to-Image Generation
  Arianna Compostella and Maria Teresa Guasti
  
• From Supervised Learning to LLMs: A Multilingual Study of Humor Detection
  Ivana Crescenzi, Jakub Jerzy Górecki, Emanuele Moscato and Debora Nozza
  
• From Weights to Representations: How Pre-Training Data Order Shapes Language Models
  Luca Dini, Lucia Domenichelli, Dominique Brunato and Felice Dell’Orletta
  
• Generating Synthetic Data for Lexical Semantic Change Detection Over Multiple Time Periods
  Davide Saponara, Nina Tahmasebi and Pierluigi Cassotti
  
• How Far Can an Agentic Framework Close the Gap between LLMs and Specialised MT for Low-Resource Languages?
  Jean Maillard, Samuel Frontull, Christopher R. Haberland, Ruth Videsott and Stefano Lusito
  
• I luoghi delle partigiane: approcci digitali per l’analisi di un corpus di scritti memorialistici resistenziali femminili
  Irene Caiazzo
  
• Identifying the Factors Underlying Differential Object Marking with Language Models: analysis and comparison of Self-Paced Reading data and Transformer-based Surprisal scores
  Giorgio Sciamanna
  
• In Short? Straight Paths over Painted Mazes: Humans and LLMs Prefer Simple (Extractive or Abstractive) Summaries to Metaphor-based ones
  Nicola Cirillo, Marzia Postorino, Antonio Schiano, Francesca Turco, Stefano Zoia and Antonio Lieto
  
• Intercepting and Traversing a Telegramsphere: Construction and Analysis of the Italian Anti-Vaccine Telegram Corpus
  Selenia Anastasi, Federico Pilati and Anita Lavorgna
  
• Investigating Idiomaticity in Word Representations for Italian
  Petra Giommarelli
  
• IPA as a Unified Representation for Multilingual ASR
  Ignazio Sorrentino, Marco Matassoni, Roberto Gretter, Daniele Falavigna and Alessio Brutti
  
• Italian Sociolinguistics and LLMs: A Study on Diaphasic, Diastratic, and Diatopic Dimensions
  Sveva Silvia Pasini, Calogero Jerik Scozzaro, Daniele P. Radicioni and Viviana Patti
  
• Know What You Reward: Criterion-Based Reward Modeling for Summarization
  Olha Khomyn, Edoardo Cecchinato and Giuseppe Riccardi
  
• Larry Caused the Car to Stop, But the Model Didn’t Notice: Transformer Blindness to the M-Heuristic
  Butnaru Stefania, Claudiu Creanga and Liviu P. Dinu
  
• Linking WordNet to the LinkEn Knowledge Base for a Linked Data Representation of Polysemous Nouns
  Lorenzo Augello and Marco Passarotti
  
• Living on the Edge: a Graph-RAG Ablation Study for Cross-Lingual Italian Legal QA
  Rachele Mignone, Davide Liga and Luigi Di Caro
  
• LLM-Based Resume–Job Matching: Experiments and Performance
  Elham Sadat Ahmadi, Daria Gordon, Carlo Strapparava and Muhammad Uzair Ul Haq
  
• Lost in Transcription: Investigating the Role of Audio Information for Video-based Visual Statement Verification
  Michela Deodati, Mohd Kaif ., Davide Testa, Giovanni Bonetta, Alessandro Lenci, Oswald Lanz, Raffaella Bernardi and Bernardo Magnini
  
• Lost in Translation? Watermark Resilience and Perplexity Degradation in Italian LLMs across Resource Tiers
  Davide Altieri, Matteo Pasotti, Elena Opreni, Alice Pietrasanta, Claudio Stamile, Marco Pasqualini, Giulia Rizzi and Elisabetta Fersini
  
• Metaphors of Domestic Violence in the Italian Political Discourse: Annotation and Experiments with the ViolMeta Dataset
  Arianna Paola Di Pascale and Rachele Sprugnoli
  
• Mining Eruptive Events from Historical Sources: A Taxonomy and LLM-Based Benchmark
  Marius Micluta-Campeanu, Costanza Marini, Alessio Palmero Aprosio and Claudia Principe
  
• Mitigating Register- and Authority-Driven Sycophancy via Cross-Lingual Activation Steering
  Gabriele Maraia, Leonardo Ranaldi and Fabio Massimo Zanzotto
  
• Morphosyntactic Patterns in Abstractive Summarization: Do LLMs Converge on a Common Italian Variety?
  Donatella Laperchia, Silvia Bernardini and Alberto Barrón-Cedeño
  
• Multilingual sentiment analysis: identifying and addressing language-dependent bias without the need for gold labels
  Elisa Di Nuovo, Mara Bernardi, Davide Colla, Nicolas Stefanovitch, Domenico Perrotta, Francesca Torti, Bertrand De Longueville and Charles Macmillan
  
• Multimodal Emotional Conflict Resolution in LLMs: A Behavioral Analysis of Incongruent Visual and Linguistic Signals
  Elham Sadat Ahmadi and Carlo Strapparava
  
• No One-Size-Fits-All Solution: Evaluating Language Identification Methods for Languages of Italy
  Debora Ciminari and Alberto Barrón-Cedeño
  
• Perspective-Aware RAG for Exploring Italian Farmers’ Opinions on the Common Agricultural Policy
  Michael Oliverio, Alessandro Mazzei, Luca Anselma, Marco Vassallo, Concetta Cardillo and Giuliano Gabrieli
  
• Plausible but Different: Comparing Human and LLM-Simulated Open-Ended Survey Responses in Italian
  Marco Rospocher, Marco Bombieri, Sofia Cramerotti, Dario Ianes and Benedetta Zagni
  
• Pretty Pictures, Shallow Parsing: The Impact of Syntactic Structure on Diffusion Models
  Chiara Ceka, Tiziano Gaddo, Marco La Marra and Roberto Zamparelli
  
• Prompt-Based Semantic Grounding via Ontology Injection for Biomedical Document Reranking
  Federico Borazio, Danilo Croce and Roberto Basili
  
• PUNder-IT: Assessing Language Models’ Sensitivity to Italian Puns
  Massimiliano Orsini and Dominique Brunato
  
• Reading Between the Clues: A Fine-Grained Analysis of Italian Crossword Solving Systems
  Cristiano Ciaccio, Asya Zanollo, Alessio Miaschi, Gabriele Sarti, Felice Dell’Orletta and Malvina Nissim
  
• Reading Comprehension Difficulty: Bridging Linguistic Profiling and National Assessment in Education
  Dominique Brunato, Marta Desimoni, Antonella Mastrogiovanni, Simonetta Montemagni, Laura Petrillo and Giulia Venturi
  
• RogueAI: A Reverse Turing Test for Detecting Licensed AI Deception in Dialogue
  Sara Candussio, Emanuele Ballarin, Lorenzo Bonin, Sandro Junior Della Rovere and Luca Bortolussi
  
• Scaling Italian Foundation Models Improves Expression More Than Knowledge: A Behavioral Model-Diffing Study of the Minerva Family on ITALIC
  Federica Fabri and Alessandro Bondielli
  
• Searching for Truth: How Reliably Does Google Surface Fact-Checked Evidence?
  Daniel Russo
  
• Semantics over Structure – on the importance of Grounding Concept Embeddings
  Davide Cavicchini, Yesun-Erdene Jargalsaikhan, Fausto Giunchiglia and Jacopo Staiano
  
• SENSEI: A Simple, Elegant, and Nimble Subtitle Editing Interface
  Mauro Cettolo, Elia Soliman, Roldano Cattoni, Matteo Negri, Luisa Bentivogli and Alessio Palmero Aprosio
  
• Soundtrack 2 My Mood: A RAG-Based LLM System for Emotionally Relatable Music Recommendations
  Filippo De Min, Debora Nozza and Arianna Muti
  
• Sparse Autoencoder Latents Are Locally Aligned with Their Explanations
  Francesco Tozzi and Alessandro Bondielli
  
• Sparse Autoencoders Reveal Representations of Cognitively Motivated Semantic Categories in LLMs
  Andrea Bruera and Roberto Dessì
  
• Temporal Linguistic Escalation Patterns in Clinical Narratives Prior to Clinical Deterioration
  Faizan Nadeem Mufti, Ajithkumar Jayasree Jayakrishna, Het Darshan Mehta, Marco Polignano, Giovanni Semeraro and Ernesto William De Luca
  
• The Accent It Believes It Hears: Accent Attribution and Bias in AI Evaluation of Human Speech
  Claudia Soria
  
• The effects of benevolent fine-tuning on the safety of the Italian Large Language Models
  Francesca Pulerà, Giulia Rizzi, Giuseppe Magazzù and Elisabetta Fersini
  
• The Linguistic History of (In)Fertility: A Diachronic and Cross-Genre Corpus Study
  Agnese Lombardi and Valentina Bambini
  
• The Teacher, the Slides, the Classroom
  Arianna Longo, Silvia Casola, Davide Martignetti, Giuseppe Noto and Marco Antonio Stranisci
  
• Topics of Resilience: Improving Historical Phase Recognition with Paleo Climate, Risk Management and Topic Modeling
  Konstantina Tripodi, Fabian Weist, Antonio Sottosanti and Fabio Celli
  
• Toward Ancient Greek Word Formation: a Digital Resource for Derivational Morphology
  Beatrice Marchesi, Silvia Zampetta, Virginia Mastellari, Luca Brigada Villa, Silvia Luraghi and Chiara Zanchi
  
• Towards Italian Concept Modeling: A Case Study on Fine-Tuning a Large Concept Model on the Italian Language
  Elio Musacchio, Lucia Siciliani and Pierpaolo Basile
  
• Training Small LLMs from Scratch: Lessons from Zagreus and Nesso
  Alessandro Ercolani, Edoardo Federici, Luisa Bentivogli and Luca Corbucci
  
• Understanding Social Workers’ Well-being through Sentiment Analysis
  Rachele Sprugnoli, Teodora Erika Uberti and Maddalena Floriana Grassi
  
• Villanova-2B: European Open Multilingual and Multimodal Language Models
  Stefano Simonazzi, Kamyar Zeinalipour, Matteo Gabburo, Daniela Baiamonte, Elena Fano and Andrea Zugarini
  
• What is my Age (and Gender) again? Profiling AI Systems Through the Lens of Gender Stereotype Detection
  Davide Testa, Gloria Comandini, Manuela Speranza, Sofia Brenna, Stefania Cavagnoli and Bernardo Magnini
  
• When Do Discourse Markers Help? A Multilingual Study of Discourse-Marker Auxiliary Training for NLP
  Guilherme Augusto Rocha de Figueiredo, Michele Luca Contalbo, Matteo Paganelli and Francesco Guerra
  
• When Does Whisper Hear a Question? Prosodic and Lexical Cues to Question-Mark Transcription in the KIParla Corpus
  Martina Simonotti and Ludovica Pannitto
  
• When Life Gives You Lemons, Make A Bitchin’ Lemonade: Evaluating Synthetic Data in the Context of Slur Reclamation
  Chiara Ferrando, Lia Draetta, Mirko Lai, Giancarlo Ruffo and Viviana Patti
  
• When Metrics Miss the Language: Machine Translation Evaluation for Piedmontese
  Yulia Panchenko
  
• Who is Responsible? Constructional Patterns of Agency in Italian Political Debate about Environment
  Elisa Chierchiello, Sveva Silvia Pasini, Arianna Paola Di Pascale, Eliana Di Palma and Cristina Bosco
  
• Who Judges the Judges? Evaluating LLM-as-a-Judge for Open-Ended Multimodal Tasks
  Davide Testa, Giovanni Bonetta, Leonardo Rigutini and Bernardo Magnini

---

Research Communications

• A Survey on Stereotype Detection in Natural Language Processing
  Alessandra Teresa Cignarella, Anastasia Giachanou and Els Lefever
  
• An experimental comparison of the most popular approaches to fake news detection
  Pietro Dell’Oglio, Alessandro Bondielli, Francesco Marcelloni and Lucia C. Passaro
  
• Differential syntactic and semantic encoding in LLMs
  Santiago Acevedo, Alessandro Laio and Marco Baroni
  
• Friend or Foe? A Computational Investigation of Semantic False Friends across Romance Languages
  Liviu P. Dinu, Ana Sabina Uban, Ioan-Bogdan Iordache, Simona Georgescu and Claudia Vlad
  
• Head Pursuit: Probing Attention Specialization in Multimodal Transformers
  Lorenzo Basile, Valentino Maiorca, Diego Doimo, Francesco Locatello and Alberto Cazzaniga
  
• LiteraryQA: Towards Effective Evaluation of Long-document Narrative QA
  Tommaso Bonomo, Luca Gioffré and Roberto Navigli
  
• Multi-LMentry: Can Multilingual LLMs Solve Elementary Tasks Across Languages?
  Luca Moroni and Roberto Navigli
  
• On automatic decipherment of lost ancient scripts relying on combinatorial optimisation and coupled simulated annealing
  Fabio Tamburini
  
• On the impact of pretraining data ordering in transformer encoder- and decoder-only language models
  Luca Dini, Lucia Domenichelli, Dominique Brunato and Felice Dell’Orletta
  
• PERSEVAL: A Framework for Perspectivist Classification Evaluation
  Soda Marem Lo, Silvia Casola, Erhan Sezerer, Valerio Basile, Franco Sansonetti, Antonio Uva and Davide Bernardi
  
• Perspectives in Play: A Multi-Perspective Approach for More Inclusive NLP Systems
  Benedetta Muscato, Lucia C. Passaro, Gizem Gezici and Fosca Giannotti
  
• RETRACEQA: Evaluating Reasoning Traces of Small Language Models in Commonsense Question Answering
  Luca Moroni, Francesco Maria Molfese, Simone Conia and Roberto Navigli
  
• The Dangerous Effects of a Frustratingly Easy LLMs Jailbreak Attack
  Marco Bombieri, Simone Paolo Ponzetto and Marco Rospocher
  
• The Narrow Gate: Localized Image-Text Communication in Native Multimodal Models
  Alessandro Pietro Serra, Francesco Ortu, Emanuele Panizon, Lucrezia Valeriani, Lorenzo Basile, Alessio Ansuini, Diego Doimo and Alberto Cazzaniga
  
• When Harry Meets Superman: The Role of The Interlocutor in Persona-Based Dialogue Generation
  Daniela Occhipinti, Marco Guerini and Malvina Nissim