FAQ
logo Uniwersytetu Jagiellońskiego w Krakowie
PL | EN

Ustawienia prywatności (cookies)

Ta strona używa ciasteczek (cookies), dzięki którym nasz serwis może działać lepiej. Aby uzyskać więcej informacji i spersonalizować swoje preferencje, kliknij „Ustawienia”. W każdej chwili możesz zmienić swoje preferencje, a także cofnąć zgodę na używanie plików cookie na poniższej stronie.
Polityka prywatności (link otworzy się w nowym oknie)
* z wyjątkiem niezbędnych plików cookies do prawidłowego działania strony oraz realizacji obowiązków prawnych administratora
Przesuwając suwak w prawo aktywujesz konkretną zgodę. Przesuwając suwak w lewo wyłączasz taką zgodę.

L’applicativo di Inclusively: una via per l’inclusione?

Data publikacji: 12.12.2025

Romanica Cracoviensia, Tom 25 (2025), Tom 25, numer 2, s. 141-149

https://doi.org/10.4467/20843917RC.25.033.22745

Autorzy

,
Tania Cerquitelli
Politecnico di Torino
, Włochy
https://orcid.org/0000-0002-9039-6226 Orcid
Kontakt z autorem
Wszystkie publikacje autora →
Matteo Berta
Politecnico di Torino
, Włochy
https://orcid.org/0009-0009-3046-0386 Orcid
Kontakt z autorem
Wszystkie publikacje autora →

Pobierz pełny tekst

Drukuj drukuj Cytuj cytuj

Tytuły

L’applicativo di Inclusively: una via per l’inclusione?

Abstrakt

The reflection on the use of language and the demand for a more inclusive communication paradigm are now at the center of public and institutional debate. In this context, new technologies and the spread of Generative Artificial Intelligence play a crucial role. Large Language Models (LLMs), the main expressions of this technology, are trained in an unsupervised way on huge amounts of uncontrolled data, thus contributing to the perpetuation of social biases. Inclusively was born with the aim of offering a generative model of inclusive language, based on supervised data produced by experts in the domain, proposing itself as an effective tool for intralingual translation from non-inclusive to inclusive language.

Bibliografia

Pobierz bibliografię

Bloch Arthur, 1980, Murphys Law Book Two: More Reasons Why Things Go Wrong!, New York: Price/Stern/Sloan.

Boroditsky Lera, 2011, “How Language Shapes Thought”, Scientific American 304: 62–65.

Greco Salvatore, La Quatra Moreno, Cagliero Luca, Cerquitelli Tania, 2025, “Towards AI-assisted Inclusive Language Writing in Italian Formal Communications”, ACM Transactions on Intelligent Systems and Technology 16 (4): 1–24, DOI: 10.1145/3729237.

CrossRef Następne

Kotek Hadas, Dockum Rikker, Sun David, 2023, “Gender Bias and Stereotypes in Large Language Models”, (in:) Proceedings of the ACM Collective Intelligence Conference (CI 23) ACM: 12–24.

Kranzberg Melvin, 1986, “Technology and History: ‘Kranzberg’s Laws’”, Technology and Culture 27 (3): 544–560.

Kumar Abhishek, Yunusov Sarfaroz, Emami Ali, 2024, “Subtle Biases Need Subtler Measures: Dual Metrics for Evaluating Representative and Affinity Bias in Large Language Models”, (in:) Proceedings of the 62nd Annual Meeting of the Association for Computational Linguistics, vol. 1: Long Papers, Bangkok: Association for Computational Linguistics, 375–392.

La Quatra Moreno, Cagliero Luca, 2023, “BART-IT: An Efficient Sequence-to-sequence Model for Italian Text Summarization”, Future Internet 15 (1), DOI: 10.3390/fi15010015.

CrossRef Następne

La Quatra Moreno, Greco Salvatore, Cagliero Luca, Tonti Michela, Dragotto Francesca, Raus Rachele, Cavagnoli Stefania, Cerquitelli Tania, 2024, “Building Foundations for Inclusiveness through Expert-annotated Data”, (in:) Proceedings of the EDBT/ICDT Workshops 2024https://ceur-ws.org/Vol-3651/DARLI-AP-3.pdf (consultato il 19 aprile 2025).

Lindsey Jacob et al., 2025, On the Biology of a Large Language Model, Transformer Circuits, https://transformer-circuits.pub/2025/attribution-graphs/biology.html (consultato il 19 aprile 2025).

Navigli Roberto, Conia Simone, Ross Björn, 2023, “Biases in Large Language Models: Origins, Inventory, and Discussion”, Journal of Data and Information Quality 15 (2): 1–21, DOI: 10.1145/3597307.

CrossRef Następne

Piergentili Andrea, Fucci Dennis, Savoldi Beatrice, Bentivogli Luisa, Negri Matteo, 2023, “Gender Neutralization for an Inclusive Machine Translation: From Theoretical Foundations to Open Challenges”, (in:) Proceedings of the First Workshop on Gender-Inclusive Translation Technologies, Tampere: European Association for Machine Translation, 71–83.

Polignano Marco, Basile Pierpaolo, de Gemmis Marco, Semeraro Giovanni, Basile Valerio, 2019, “AIBERTo: Italian BERT Language Understanding Model for NLP Challenging Tasks Based on Tweets”, (in:) Proceedings of the 6th Italian Conference on Computational Linguistics (CLiC-it 19), Bari: CEUR Workshop Proceedings, 312–317.

Sarti Gabriele, Nissim Malvina, 2024, “IT5: Text-to-text Pretraining for Italian Language Understanding and Generation”, (in:) Proceedings of the 2024 Joint International Conference on Computational Linguistics, Language Resources and Evaluation (LREC-COLING 2024), Torino: ELRA and ICCL, 9422–9433.

Shrawgi Hari, Rath Prasanjit, Singhal Tushar, Dandapat Sandipan, 2024, “Uncovering Stereotypes in Large Language Models: A task Complexity-based Approach”, (in:) Proceedings of the 18th Conference of the European Chapter of the Association for Computational Linguistics, vol. 1: Long Papers, St. Julian’s: Association for Computational Linguistics, 1841–1857.

Ta Vivian, Boyd Ryan L., Seraj Sahar, Keller Anne, Griffith Caroline, Loggarakis Alexia, Medema Lael, 2022, “An Inclusive, Real-world Investigation of Persuasion in Language and Verbal Behavior”, Journal of Computational Social Science 5: 883–903.

Wittgenstein Ludwig, 1922, Tractatus Logico-Philosophicus, London: Kegan Paul.

Informacje

Informacje: Romanica Cracoviensia, Tom 25 (2025), Tom 25, numer 2, s. 141-149

Typ artykułu: Oryginalny artykuł naukowy

Tytuły:

Włoski: L’applicativo di Inclusively: una via per l’inclusione?
Angielski: The Inclusively app: A path for inclusion?

Publikacja: 12.12.2025

Status artykułu: Otwarte __T_UNLOCK

Licencja: CC BY 4.0  ikona licencji

Finansowanie artykułu:

Questo studio è stato realizzato nell’ambito del progetto “E-MIMIC: Empowering Multilingual Inclusive Communication” (Nr. 2022WEFCFP), finanziato dal Ministero dell’Università e della Ricerca – con il programma PRIN 2022 (D.D. 104 – 02/02/2022).

Udział procentowy autorów:

Tania Cerquitelli (Autor) - 50%
Matteo Berta (Autor) - 50%

Informacje o autorze:

Tania Cerquitelli is Full Professor at the Department of Control and Computer Engineering at Politecnico di Torino and serves in an affiliated role to the Vice-Rector for Society, Community, and the Implementation of the Program. Her research focuses on technological and algorithmic innovation for data valorization. Specifically, she develops algorithms and methods for responsible artificial intelligence, aligned with the EU AI Act, designed to be accurate, human-supervised, explainable, and adaptive, with strategies to mitigate bias and risk. She is the principal investigator of the E-MIMIC project (Empowering Multilingual Inclusive comMunICation, 2023–2025), funded by the Italian Ministry of University and Research (MUR) under the PRIN-22 call. The project aims to promote inclusive communication through deep learning algorithms for natural language processing. She has led over 20 research and technology transfer projects in machine learning, funded by public institutions and private companies, and serves on the editorial boards of several international journals.

Matteo Berta is a Research Fellow in Data Science at Politecnico di Torino, Italy, with a background in Computer Engineering. Guided by the principle that “the future belongs to the curious”, his interests span a wide range of disciplines, from the humanities to the sciences. His current research focuses on natural language processing, particularly issues related to fairness, stereotype detection, and time-series forecasting. Outside academia, he is passionate about cycling, literature, music, and photography. His curiosity fuels a continuous pursuit of knowledge, creativity, and innovation across diverse fields.

Korekty artykułu:

-

Języki publikacji:

Angielski, Włoski

Liczba wyświetleń: 276

Liczba pobrań: 391

L’applicativo di Inclusively: una via per l’inclusione?

L’applicativo di Inclusively: una via per l’inclusione?

cytuj

Pobierz PDF Pobierz

pobierz pliki

RIS BIB ENDNOTE