Check out the program of the AIxIA conference at this link!
DAY 1: Wednesday, October 7th, 2026
📍 Accademia di Belle Arti
Piazza San Francesco al Prato, 5, 06123 Perugia PG
MORNING SESSION (10:30 – 13:00)
>> 10:30-10:40 – OPENING DAY 1
Welcome to NL4AI 2026
>> 10:40-11:35 – KEYNOTE SPEAKER
Marco Turchi – Head of the Machine Translation group at Zoom
Title: Beyond Models: The Impact of AI on NLP through the Lens of Machine Translation
>> 11:35-12:53 – ORAL SESSION 1
Evaluation, Bias and Clinical Applications
11:35-11:48 – Martina Miliani and Chiara Manno
Same Job, Different Gender: Uncovering LLM Stereotypes via Counterfactual Biographies
11:48-12:01 – Samuele D’Avenia, Timothy Meinert, Marta Marchiori Manerba, Valerio Basile and Anna Koufakou
Does Perspective Matter? Annotation Role Alignment in Emotion Recognition
12:01-12:14 – Salvatore Ricciardi, Simone Scaboro, Beatrice Turano, Cristina Giannone, Andrea Favalli and Raniero Romagnoli
Who Judges the Judges? A Multilingual Validity Probe for Complementary LLM Evaluation Metrics
12:14-12:27 – Michał Tomasz Godziszewski, Mateusz Galiński, Ryszard Tuora, Mateusz Czyżnikiewicz, Tomasz Ziętkiewicz, Adam Kozakiewicz and Cristina Cornelio
Election-Based Evaluation Methods for NLG-Systems
12:27-12:40 – Irene Siragusa, Roberto Pirrone
Evaluating Clinical Coding Alignment Across ICD-9, ICD-10, and ICD-11 with a Semantic Retrieval Approach
12:40-12:53 – Claire Lawand, Haeeul Hwang, Jalal Al-Tamimi, Valeria Lucarini, Mahshid Hozhabr, Deok-Hee Kim-Dufor, Christophe Lemey, Julien Desclés and Motasem Alrahabi
The Prosody of Clinical Speech: Annotation and Computational Analysis of Psychiatric Interviews
13:00-14:30 – Lunch Break
AFTERNOON SESSION (14:30-16:30)
>> 14:30-15:25 – KEYNOTE SPEAKER
Lorenzo De Mattei – Co-founder and CPTO at APTUS.AI
Title: Back to the Future: From Models Back to Systems in NLP
>> 15:25-15:50 ORAL SESSION 2
Misinformation and Harmful Content
15:25-15:37 – Endang Wahyu Pamungkas, Maryam Maryam, Rona Rizkhy Bunga Chasana, Dade Nurjanah and Ikhlasul Amal
Static or Dynamic? Exploring Cross-Lingual Few-Shot Transfer for Indonesian Misogynistic Meme Detection
15:37-15:50 – Alessandro Pucci, Danilo Croce and Roberto Basili
Scale or Specialization? Evaluating Language Models for Italian Fact Verification
>> 15:50:16:30 ROUNDTABLE
Is there Life after LLMs?
DAY 2: Thursday, October 8th, 2026
📍 Accademia di Belle Arti
Piazza San Francesco al Prato, 5, 06123 Perugia PG
AFTERNOON SESSION (14:30-16:15)
>> 14:30-16:05 ORAL SESSION 3
Robustness, Reasoning and Low-Resource NLP
14:30-14:43 – Francesco Colace, Sabrina Galano, Angelo Lorusso, Domenico Santaniello, Giusy Strollo and Annapia Tanzola
Machine Translation of Low-Resource Historical Languages: A Comparative Study of LLMs on Old Occitan
14:43-14:56 – Elena Opreni, Alice Pietrasanta, Davide Altieri, Giulia Rizzi and Elisabetta Fersini
Evaluating Watermark Resilience in Italian Large Language Models Under Back-Translation Attacks
14:56-15:09 – Stefano Scotta, Giulio Alfarano and Alberto Messina
Comparing Generated Pseudo-Labeled News Datasets: A Retrieval-and-Verification Approach
15:09-15:22 – Mouhamed Mbaye
Known but Unreachable: A Diagnostic Evaluation of Wolof Entity Linking against Wikidata
15:22-15:35 – Serena Auriemma, Martina Miliani, Alessandro Bondielli, Lucia C. Passaro and Alessandro Lenci
Right for the Right Reasons? A Chain-of-Thought Analysis of Commonsense Causal Reasoning
15:35-15:48 – Goshgar Ismayilov
MAWGE: A Parametric and Synthetic Multi-Agentic Workflow Generator Library
15:48-16:01 – Ejdis Gjinika, Luca Putelli, Ivan Serina and Alfonso Emilio Gerevini
Exploring the Relationship between Sparse Autoencoder Activations and Sentence Complexity
>> 16:05-16:15 Closing remarks