Italian Transformers Under the Linguistic Lens

Alessio Miaschi; Gabriele Sarti; Dominique Brunato; Felice Dell'Orletta; Giulia Venturi

Italian Transformers Under the Linguistic Lens

Alessio Miaschi^*, Gabriele Sarti, Dominique Brunato, Felice Dell'Orletta, Giulia Venturi

^*Bijbehorende auteur voor dit werk

Onderzoeksoutput › Academic › peer review

Samenvatting

In this paper we present an in-depth investigation of the linguistic knowledge encoded by the transformer models currently available for the Italian language. In particular, we investigate whether and how using different architectures of probing models affects the performance of Italian transformers in encoding a wide spectrum of linguistic features. Moreover, we explore how this implicit knowledge varies according to different textual genres.

Originele taal-2	English
Titel	Proceedings of the Seventh Italian Conference on Computational Linguistics (CLiC-it 2020)
Redacteuren	Johanna Monti, Felice Dell'Orletta, Fabio Tamburini
Uitgeverij	CEUR Workshop Proceedings (CEUR-WS.org)
Status	Published - 1-mrt.-2021
Extern gepubliceerd	Ja
Evenement	Italian Conference on Computational Linguistics 2020 - Bologna, Italy Duur: 1-mrt.-2021 → 3-mrt.-2021

Conference

Conference	Italian Conference on Computational Linguistics 2020
Verkorte titel	CLiC-it 2020
Land/Regio	Italy
Stad	Bologna
Periode	01/03/2021 → 03/03/2021

Toegang tot document

http://ceur-ws.org/Vol-2769/paper_56.pdfLicentie: CC BY

Handle.net

Permanente koppeling

Andere bestanden en links

http://ceur-ws.org/Vol-2769/paper_56.pdf

1 Article

Probing Linguistic Knowledge in Italian Neural Language Models across Language Varieties
Miaschi, A., Sarti, G., Brunato, D., Dell’Orletta, F. & Venturi, G., jul.-2022, In: Italilan Journal of Computational Linguistics. 8, 1, blz. 25-44 20 blz., 2.
Onderzoeksoutput › Academic › peer review

Open Access
Bestand

23 Downloads (Pure)

Citeer dit

@inproceedings{9012589433fe4b6285107c22ae2db77d,

title = "Italian Transformers Under the Linguistic Lens",

abstract = "In this paper we present an in-depth investigation of the linguistic knowledge encoded by the transformer models currently available for the Italian language. In particular, we investigate whether and how using different architectures of probing models affects the performance of Italian transformers in encoding a wide spectrum of linguistic features. Moreover, we explore how this implicit knowledge varies according to different textual genres.",

keywords = "natural language processing, deep learning, probing task, interpretability, italian language, neural language models, transformers",

author = "Alessio Miaschi and Gabriele Sarti and Dominique Brunato and Felice Dell'Orletta and Giulia Venturi",

year = "2021",

month = mar,

day = "1",

language = "English",

editor = "Johanna Monti and Felice Dell'Orletta and Fabio Tamburini",

booktitle = "Proceedings of the Seventh Italian Conference on Computational Linguistics (CLiC-it 2020)",

publisher = "CEUR Workshop Proceedings (CEUR-WS.org)",

note = "Italian Conference on Computational Linguistics 2020, CLiC-it 2020 ; Conference date: 01-03-2021 Through 03-03-2021",

}

Miaschi, A, Sarti, G, Brunato, D, Dell'Orletta, F & Venturi, G 2021, Italian Transformers Under the Linguistic Lens. in J Monti, F Dell'Orletta & F Tamburini (redactie), Proceedings of the Seventh Italian Conference on Computational Linguistics (CLiC-it 2020). CEUR Workshop Proceedings (CEUR-WS.org), Italian Conference on Computational Linguistics 2020, Bologna, Italy, 01/03/2021. <http://ceur-ws.org/Vol-2769/paper_56.pdf>

TY - GEN

T1 - Italian Transformers Under the Linguistic Lens

AU - Miaschi, Alessio

AU - Sarti, Gabriele

AU - Brunato, Dominique

AU - Dell'Orletta, Felice

AU - Venturi, Giulia

PY - 2021/3/1

Y1 - 2021/3/1

N2 - In this paper we present an in-depth investigation of the linguistic knowledge encoded by the transformer models currently available for the Italian language. In particular, we investigate whether and how using different architectures of probing models affects the performance of Italian transformers in encoding a wide spectrum of linguistic features. Moreover, we explore how this implicit knowledge varies according to different textual genres.

AB - In this paper we present an in-depth investigation of the linguistic knowledge encoded by the transformer models currently available for the Italian language. In particular, we investigate whether and how using different architectures of probing models affects the performance of Italian transformers in encoding a wide spectrum of linguistic features. Moreover, we explore how this implicit knowledge varies according to different textual genres.

KW - natural language processing

KW - deep learning

KW - probing task

KW - interpretability

KW - italian language

KW - neural language models

KW - transformers

UR - http://ceur-ws.org/Vol-2769/paper_56.pdf

M3 - Conference contribution

BT - Proceedings of the Seventh Italian Conference on Computational Linguistics (CLiC-it 2020)

A2 - Monti, Johanna

A2 - Dell'Orletta, Felice

A2 - Tamburini, Fabio

PB - CEUR Workshop Proceedings (CEUR-WS.org)

T2 - Italian Conference on Computational Linguistics 2020

Y2 - 1 March 2021 through 3 March 2021

ER -

Italian Transformers Under the Linguistic Lens

Samenvatting

Conference

Toegang tot document

Handle.net

Andere bestanden en links

Onderzoekersoutput

Probing Linguistic Knowledge in Italian Neural Language Models across Language Varieties

Citeer dit