AILAB-Udine@SMM4H’22: Limits of Transformers and BERT Ensembles

Beatrice Portelli; Simone Scaboro; Emmanuele Chersoni; Enrico Santus; Giuseppe Serra

AILAB-Udine@SMM4H’22: Limits of Transformers and BERT Ensembles

Beatrice Portelli, Simone Scaboro, Emmanuele Chersoni, Enrico Santus, Giuseppe Serra

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

Abstract

This paper describes the models developed by the AILAB-Udine team for the SMM4H’22 Shared Task. We explored the limits of Transformer based models on text classification, entity extraction and entity normalization, tackling Tasks 1, 2, 5, 6 and 10. The main takeaways we got from participating in different tasks are: the overwhelming positive effects of combining different architectures when using ensemble learning, and the great potential of generative models for term normalization.

Original language	English
Title of host publication	Proceedings of The Seventh Workshop on Social Media Mining for Health Applications, Workshop & Shared Task (SMM4H 2022)
Editors	Graciela Gonzalez-Hernandez, Davy Weissenbacher
Pages	130–134
Publication status	Published - Oct 2022
Event	The 29th International Conference on Computational Linguistics - Gyeongju, Korea, Democratic People's Republic of Duration: 12 Oct 2022 → 17 Oct 2022 http://coling2022.org/

Conference

Conference	The 29th International Conference on Computational Linguistics
Abbreviated title	COLING2022
Country/Territory	Korea, Democratic People's Republic of
City	Gyeongju
Period	12/10/22 → 17/10/22
Internet address	http://coling2022.org/

More information

Cite this

@inproceedings{8a536aa2a3d141389ea6f6c2a5f75c6c,

title = "AILAB-Udine@SMM4H{\textquoteright}22: Limits of Transformers and BERT Ensembles",

abstract = "This paper describes the models developed by the AILAB-Udine team for the SMM4H{\textquoteright}22 Shared Task. We explored the limits of Transformer based models on text classification, entity extraction and entity normalization, tackling Tasks 1, 2, 5, 6 and 10. The main takeaways we got from participating in different tasks are: the overwhelming positive effects of combining different architectures when using ensemble learning, and the great potential of generative models for term normalization.",

author = "Beatrice Portelli and Simone Scaboro and Emmanuele Chersoni and Enrico Santus and Giuseppe Serra",

year = "2022",

month = oct,

language = "English",

isbn = "2951-2093",

pages = "130–134",

editor = "Graciela Gonzalez-Hernandez and Davy Weissenbacher",

booktitle = "Proceedings of The Seventh Workshop on Social Media Mining for Health Applications, Workshop & Shared Task (SMM4H 2022)",

note = "The 29th International Conference on Computational Linguistics, COLING2022 ; Conference date: 12-10-2022 Through 17-10-2022",

url = "http://coling2022.org/",

}

Portelli, B, Scaboro, S, Chersoni, E, Santus, E & Serra, G 2022, AILAB-Udine@SMM4H’22: Limits of Transformers and BERT Ensembles. in G Gonzalez-Hernandez & D Weissenbacher (eds), Proceedings of The Seventh Workshop on Social Media Mining for Health Applications, Workshop & Shared Task (SMM4H 2022). pp. 130–134, The 29th International Conference on Computational Linguistics, Gyeongju, Korea, Democratic People's Republic of, 12/10/22. <https://arxiv.org/pdf/2209.03452.pdf>

AILAB-Udine@SMM4H’22: Limits of Transformers and BERT Ensembles. / Portelli, Beatrice; Scaboro, Simone; Chersoni, Emmanuele et al.
Proceedings of The Seventh Workshop on Social Media Mining for Health Applications, Workshop & Shared Task (SMM4H 2022). ed. / Graciela Gonzalez-Hernandez; Davy Weissenbacher. 2022. p. 130–134.

Research output: Chapter in book / Conference proceeding › Conference article published in proceeding or book › Academic research › peer-review

TY - GEN

T1 - AILAB-Udine@SMM4H’22: Limits of Transformers and BERT Ensembles

AU - Portelli, Beatrice

AU - Scaboro, Simone

AU - Chersoni, Emmanuele

AU - Santus, Enrico

AU - Serra, Giuseppe

PY - 2022/10

Y1 - 2022/10

N2 - This paper describes the models developed by the AILAB-Udine team for the SMM4H’22 Shared Task. We explored the limits of Transformer based models on text classification, entity extraction and entity normalization, tackling Tasks 1, 2, 5, 6 and 10. The main takeaways we got from participating in different tasks are: the overwhelming positive effects of combining different architectures when using ensemble learning, and the great potential of generative models for term normalization.

AB - This paper describes the models developed by the AILAB-Udine team for the SMM4H’22 Shared Task. We explored the limits of Transformer based models on text classification, entity extraction and entity normalization, tackling Tasks 1, 2, 5, 6 and 10. The main takeaways we got from participating in different tasks are: the overwhelming positive effects of combining different architectures when using ensemble learning, and the great potential of generative models for term normalization.

M3 - Conference article published in proceeding or book

SN - 2951-2093

SP - 130

EP - 134

BT - Proceedings of The Seventh Workshop on Social Media Mining for Health Applications, Workshop & Shared Task (SMM4H 2022)

A2 - Gonzalez-Hernandez, Graciela

A2 - Weissenbacher, Davy

T2 - The 29th International Conference on Computational Linguistics

Y2 - 12 October 2022 through 17 October 2022

ER -

AILAB-Udine@SMM4H’22: Limits of Transformers and BERT Ensembles

Abstract

Conference

More information

Fingerprint

Cite this