EVALITA 2026
9th Evaluation Campaign of Natural Language Processing and Speech Tools for Italian

Table of Contents

• Preface
  Summary: There were 57 papers submitted for peer-review and accepted to this workshop.
• EVALITA 2026: Overview of the 9th evaluation campaign of natural language processing and speech tools for italian
  Francesco Cotugno, Alessio Miaschi, Alessio Palmero Aprosio, Giulia Rambelli, Lucia Siciliani and Marco Antonio Stranisci

Keynote Talk

• Tu chiamale se vuoi emozioni: The impact of emotions on disinformation and sexism identification
  Paolo Rosso

Multi-Track System

• Kenji-Endo: a BabyLM @EVALITA
  Calogero Jerik Scozzaro, Matteo Rinaldi, Gianluca Mittone and Marco Antonio Stranisci

Track "Authorship Analysis"

Desegma-IT - Detection and Segmentation of Machine Generated Text in Italian

• DeSegMa-IT at EVALITA 2026: Overview of the ”Detection and Segmentation of Machine generated texts in Italian”
  Giovanni Puccetti, Andrea Esuli and Andrea Pedrotti
• Gradient Descenders at DeSegMa-IT: Leveraging Monolingual Transformer for LLM-Generated Text Detection and Boundary Identification
  Tran Phuoc Thanh Nhan, Bui Hong Son and Dang Van Thin
• MINDS at DeSegMa-IT: Detecting Human-LLM Authorship Switches via Token-Level Classification
  Flavio Giobergia
• Nicla at DeSegMa-IT: DistilBERT for MGT Detection and LightGBM Regressor for HMT Segmentation
  Nicola Auletta
• Stochastic Gradient Descenders at DeSegMa-IT: Instruction-Tuned LLM and Token Classification for MGT Detection and Boundary Localization
  Huynh Nguyen Phu, Bui Hong Son and Dang Van Thin
• UniTor at DeSegMa-It: Analyzing Supervision and Encoder Representations for Italian Machine-Generated Text Detection
  Federico Boraziio, Giacomo De Luca, Daniele Pasquini, Danilo Croce and Roberto Basili

Track "Computational Ethics"

GSI:detect – Detecting Gender Stereotypes in Italian

• GSI:detect at EVALITA 2026: Overview o the Task on Detecting Gender Stereotypes in Italian
  Gloria Comandini, Manuela Speranza, Sofia Brenna, Davide Testa, Stefania Cavagnoli and Bernardo Magnini
• DIAG-Sapienza at GSI:detect: Joint Detection and Classification of Gender Stereotypes with Structured Prompting and Fine-Tuning
  Olga E. Sorokoletova, Emanuele Musumeci and Daniele Nardi
• Festa at GSI:detect: In-Context Learning vs. Fine-Tuning for Gender Stereotype Detection
  Donato Festa
• MINDS at GSI:detect: From Logits to Degrees of Agreement in Gender Stereotype Detection with LLMs
  Flavio Giobergia
• StereoBusters at GSI:detect: LLM-Based Detection and Human Qualitative Analysis of Gender Stereotypes in Italian Short Text
  Salvatore Greco, Moreno La Quatra, Marta Marchiori Manerba, Ricardo Muñoz Sánchez and Alessandra Teresa Cignarella
• Team Prisma at GSI:detect: Comparing PB&J Persona-Based and Few-Shot Approaches to Gender Stereotype Detection at EVALITA 2026
  Claudia Zaghi
• Tiz at GSI:detect: Modeling Gender Stereotype Detection as Multi-Category Gender Stereotype Scoring
  Tiziano LaBruna

MultiPRIDE: Multilingual Automatic Detection of Reclamation of Slurs in the LGBTQ+ Context

• MultiPRIDE at EVALITA 2026: Overview of the Multilingual Automatic Detection of Slur Reclamation in the LGBTQ+ Context Task
  Chiara Ferrando, Lia Draetta, Marco Madeddu, Mae Sosto, Viviana Patti, Paolo Rosso, Cristina Bosco, Jacinto Mata and Estrella Gualda
• AIWizards at MULTIPRIDE: A Hierarchical Approach to Slur Reclamation Detection
  Luca Tedeschini and Matteo Fasulo
• AVAHI at MultiPRIDE: Multilingual Reclaimed Language Detection via Knowledge Graphs and Retrieval-Augmented Generation
  Tania Alcántara, Omar Garcia-Vazquez, Jose A. Torres-León, Marco Cardoso-Moreno, Diana Jimenez and Luis Moreno-Mendieta
• Challenger at MultiPRIDE: Is It Hate Speech or Reclaimed?
  Hadi Bayrami Asl Tekanlou, Mahdi Bakhtiyarzade and Jafar Razmara
• DataSummit at MultiPRIDE: Context-Aware Multilingual Detection of Slur Reclamation in LGBTQ+ Contexts
  Federico Dingeo and Marco Viviani
• Ghavidel-Rajabi at MultiPRIDE: Identity, Toxicity, or Complexity? A Language-Specific Feature Selection Approach to Reclamatory Intent Detection
  Houman Rajabi, Fatemeh Ghavidel and Kourosh Ghahremani
• GRUPPETTOZZO at MultiPRIDE 2026: Detecting LGBTQ+ Reclamatory Intent via Context-Aware Transformers
  Federico Traina, Alessandro Santoro, Gabriele Greco, Irene Siragusa and Roberto Pirrone
• HateItOff at MultiPRIDE: Linguistic and Sentiment Cues in Reclaimed LGBTQ+ Slur Detection
  Greta Damo and Nicolás Benjamín Ocampo
• I2C At MultiPRIDE 2026: Transformers And Generative LLMs For Multilingual Reclamation Classification
  Laura Vázquez-Ramos, Jacinto Mata-Vázquez and Victoria Pachón-Álvarez
• INFOTEC-NLP at MultiPRIDE: Region-Aware Transformer Models for Spanish Reclaimed Language Classification
  Jorge Gleaves and Guillermo Ruiz
• KIT-TIP-NLP at MultiPride: Continual Learning with Multilingual Foundation Model
  Barathi Ganesh HB, Michal Ptaszynski, Rene Melendez and Juuso Eronen
• LlaNa at MultiPRIDE 2026: A Fine-Tuning Approach with Cross-Lingual Augmentation
  Alessio Mercurio, Giorgio Talluto, Irene Siragusa and Roberto Pirrone
• MilaNLP at MultiPRIDE: Evaluating Lexical, Transformer, and Retrieval-Augmented Models for Reclaimed Language
  Ivana Crescenzi, Arianna Muti and Debora Nozza
• NamDang at MultiPride@EVALITA 2026: Multilingual Classification of Reclaimed Language in LGBTQ+ Discourse using Transformer-based Models
  Nam Dang, Vo Tuan Kiet and Dang Van Thin
• NetGuardAI at MultiPRIDE: Multilingual Detection of Reclaimed Language
  Noe Come Jacques Le Pollotec, Elena Simona Apostol and Ciprian Octavian Truică
• SaFe Tweets @ MultiPRIDE: From Multi-Model Ensembles to Expert LLMs for Reappropriation Detection
  Sara Visconti and Federica Manzi
• The Hate Busters at MultiPRIDE: Automatic Identification of Reappropriated Slurs in Multilingual LGBTQ+ Discourse
  Aurora Cimminelli, Giulia Corvino, Camilla Gentili and Marco Viviani
• UniBO-FICLIT at MultiPRIDE: Fine-Tuning an ELECTRA-Based Model for the Detection of Italian Reclaimed Slurs
  Simone Casazza

Track "Emerging Tasks"

Cruciverb-IT: Crossword Solving

• Cruciverb-IT at EVALITA 2026: Overview of the Crossword Solving in Italian Task
  Cristian Ciaccio, Gabriele Sarti, Alessio Miaschi, Felice Dell'Orletta and Malvina Nissim
• AC/DG at Cruciverb-IT: Retrieval-Based Approaches for Italian Crossword Clue Answering
  Ali Yassine, Claudio Savelli, Davide Napolitano, Giuseppe Gallipoli, Luca Cagliero and Elena Baralis
• FFT-UniBa at Cruciverb-IT: Special Length Tokens and CSP for Italian Crossword Solving
  Andrea Porcelli, Filippo Di Gravina, Emanuele Fontana, Mattia Curri and Francesco Damiano Di Gregorio
• MINDS at Cruciverb-IT: Solving Italian Crossword Clues with Masked Language Models and Candidate Pooling
  Flavio Giobergia
• UNIBA at Cruciverb-IT: Solving Italian Crosswords with Encoder–Decoder Models and Beam Search
  Pierpaolo Basile
• UniTor at Cruciverb-IT: Retrieval-Augmented Two-Step Reasoning for Italian Crossword Clue Answering
  Andriy Shcherbakov, Danilo Croce and Roberto Basili

SVELA: Selective Verification of Erasure from LLM Answers

• SVELA at EVALITA 2026: Overview of the Selective Verification of Erasure from LLM Answers Task
  Claudio Savelli, Moreno La Quatra, Alkis Koudounas and Flavio Giobergia
• Eraserhead at SVELA: Detecting LLM Forgetting via Logit-Space Statistics
  Matteo Berta and Tiziana Cerquitelli
• Ita-Lib at SVELA: Detecting the Forgotten — Representation-Based Approach for Verifying Machine Unlearning
  Ali Yassine, Hadi Ibrahim and Luca Cagliero
• MALTO at SVELA: A Specific-Attention-Head Approach for Membership Inference Attacks in LLMs Unlearning Evaluation
  Evren A. Munis, Mattia Sabato, Erfan Bayat and Andrea Lolli
• priyam_saha17 at SVELA: A Feature-Centric Pipeline for Verifying Selective Forgetting in Large Language Models
  Priyam Saha

Track "Multimodality"

EVWSD-ITA: Enhanced Visual Word Sense Disambiguation for Italian

• EVWSD-ITA at EVALITA 2026: Overview of the Enhanced Visual Word Sense Disambiguation for Italian Task
  Elio Musacchio, Lucia Siciliani, Pierpaolo Basile and Giovanni Semeraro
• UniTor at EVWSD-ITA: Zero-Shot Visual Word Sense Disambiguation via Visual Question-Answering
  Claudio D. Hromei, Antonio Scaiella, Danilo Croce and Roberto Basili

Track "New Challenges in Long-Standing Tasks"

ATE-IT: Automatic Term Extraction – Italian Testbed

• ATE-IT at EVALITA 2026: Overview of the Automatic Term Extraction Italian Testbed Task
  Nicola Cirillo, Giorgio Maria Di Nunzio and Federica Vezzani
• DSBKTE at ATE-IT: From Token Classification to Zero-Shot Generation: Two Approaches to Italian ATE at EVALITA 2026
  Danil Smirnov, Bita Khashechian and Giorgio Maria Di Nunzio
• JTTE at ATE-IT: A CRF Model with Contextual Embeddings
  Juliette Tonneau and Giorgio Maria Di Nunzio
• MKTE at ATE-IT: CRF-Based Term Extraction for Italian Waste Management Documents
  Minseok Kim and Giorgio Maria Di Nunzio
• OATE at ATE-IT: A Hybrid Approach for Automatic Terminology Extraction in Italian
  Omar Arab and Giorgio Maria Di Nunzio
• Peacemaker at ATE-IT: Automatic term extraction from Italian text for waste management data using encoder model
  Mahdi Bakhtiyarzadeh, Hadi Bayrami Asl Tekanlou and Jafar Razmara
• SMTE at ATE-IT: Ensemble Term Extraction with Italian BERT, spaCy, and Vocabulary-Based Filtering
  Sofia Maule and Giorgio Maria Di Nunzio
• TermNinjas at ATE-IT: Overview of the Term Extraction and Term Variants Clustering Task
  Zunaira Hasnain and Fashad Ahmed Siddqu
• TrietNLP at ATE-IT: A Hybrid Pipeline for Italian Waste Management Terminology Analysis
  Nguyen Minh Triet and Dang Van Thin
• VVTE at ATE-IT: From Candidates to Terms: Hybrid Italian ATE with Dependency Heuristics, Gemini, and Random Forest Filtering
  Valentine G. L. Vandervoort and Giorgio Maria Di Nunzio

PFB: Prometeia Financial Benchmark – Benchmarking Language Models in the Financial Domain

• PFB at EVALITA 2026: Overview of the Prometeia Financial Benchmark
  Alessandro P. Bardelli, Tolga Çekiç, Irem Demirtaş, Michele Filannino, Simona Scala, Andrea Galassi, Gianmarco Pappacoda and Paolo Torroni
• AMSN at PFB: Overview of the Automated Multilingual Economics Question Answering Task
  Ali Saleh Mohammadabad and Mehregan Nazarmohsenifakori
• UniTor at PFB: Constrained Agentic Prompting and Self-Consistency for Financial Multi-Choice QA
  Federico Borazio, Seyed Alireza Mousavian Anaraki, Shahid Iqbal Rai, Danilo Croce and Roberto Basili

Track "Pragmatics"

FadeIT: Fallacy Detection in Italian Social Media Texts

• FadeIT at EVALITA 2026: Overview of the Fallacy Detection in Italian Social Media Texts Task
  Alan Ramponi and Sara Tonelli
• Label at FadeIT: Fallacy-Aware LLM Reasoning for Score-Based Classification
  Tiziano Labruna and Eleni Papadopulos
• MALTO at FadeIT: A BERT-Based System for Multi-Label Fallacy Detection in Italian Social Media
  Matin Salami, Luca M. Rodia, Vladimir Schiau1, Evren A. Munis, Claudio Savelli and Flavio Giobergia
• PuDy at FadeIT: Enhancing Fine-Grained Fallacy Detection with Hierarchy-Aware Training and Contextual Enrichment
  Duy Dang Phu, Son Bui Hong and Thin Van Dang
• RBG-AI at FadeIT: Prompted LLMs with Label Abstraction for Logical Fallacy Detection
  Meenakshi, Jairam R, Reshma U, Barathi Ganesh HB and Michal Ptaszynski
• TIGRO at FadeIT: E Pluribus Unum – A Multi-task Approach to Fallacy Detection and Span Identification
  Stefano Atzeni, Gabriele Sarti, Tommaso Caselli and Malvina Nissim
• Unica at FadeIT: Adapting Large Language Models to Fallacy Identification in Social Networks
  Matteo Fenu and Maurizio Atzori

IMPOLS: IMplicit contents in POLitical Speech

• IMPOLS at EVALITA 2026: Overview of the IMPOLS Task
  Lorenzo Gregori, Walter Paci and Valentina Saccone
• RES2 at IMPOLS 2026: Implicit Content Classification via Span-Awareness, Hierarchical Multi-Task Learning and Synthetic Data
  Salvatore Spezia, Salvatore Ferrara, Roberto Dioguardi, Ernesto Davì, Irene Siragusa and Roberto Pirrone

CALAMITA

• INDAQA2 - A Large Italian Narrative QA Benchmark: A CALAMITA 2026 Challenge
  Luca Gioffré, Luca Moroni, Alberte Fernández-Castro, Elena Marafatto, Giacomo Garufi, and Roberto Navigli
• ProverbIT - Easy to complete, hard to choose: A CALAMITA Challenge
  Enrico Mensa, Lorenzo Zane, Calogero J. Scozzaro, Matteo Delsanto, Tommaso Milani, and Daniele P. Radicioni