Powered by OpenAIRE graph
Found an issue? Give us feedback
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/ https://aclanthology...arrow_drop_down
image/svg+xml art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos Open Access logo, converted into svg, designed by PLoS. This version with transparent background. http://commons.wikimedia.org/wiki/File:Open_Access_logo_PLoS_white.svg art designer at PLoS, modified by Wikipedia users Nina, Beao, JakobVoss, and AnonMoos http://www.plos.org/
https://aclanthology.org/2021....
Article
License: CC BY
Data sources: UnpayWall
https://doi.org/10.18653/v1/20...
Article . 2021 . Peer-reviewed
Data sources: Crossref
https://dx.doi.org/10.60692/3q...
Other literature type . 2021
Data sources: Datacite
https://dx.doi.org/10.60692/dq...
Other literature type . 2021
Data sources: Datacite
DBLP
Conference object
Data sources: DBLP
versions View all 4 versions
addClaim

This Research product is the result of merged Research products in OpenAIRE.

You have already added 0 works in your ORCID record related to the merged Research product.

YoungSheldon at SemEval-2021 Task 7: Fine-tuning Is All You Need

YoungSheldon في SemEval -2021 المهمة 7: الضبط الدقيق هو كل ما تحتاجه
Authors: Mayukh Sharma; Ilanthenral Kandasamy; W. B. Vasantha;

YoungSheldon at SemEval-2021 Task 7: Fine-tuning Is All You Need

Abstract

Dans cet article, nous décrivons notre système utilisé pour la tâche 7 de SemEval 2021 : HaHackathon : Détection et évaluation de l'humour et de l'offense. Nous avons utilisé une approche d'ajustement simple utilisant différents modèles de langage préformés (PLM) pour évaluer leurs performances pour la détection de l'humour et de l'offense. Pour les tâches de régression, nous avons fait la moyenne des scores de différents modèles conduisant à de meilleures performances que les modèles originaux. Nous avons participé à toutes les sous-tâches. Notre système le plus performant a été classé 4 dans la sous-tâche 1-b, 8 dans la sous-tâche 1-c, 12 dans la sous-tâche 2, et a bien performé dans la sous-tâche 1-a. Nous montrons en outre des résultats complets en utilisant différents modèles de langage préformés qui aideront comme bases de référence pour les travaux futurs.

En este documento, describimos nuestro sistema utilizado para la tarea 7 de SemEval 2021: HaHackathon: Detección y calificación del humor y la ofensa. Utilizamos un enfoque de ajuste fino simple utilizando diferentes modelos de lenguaje preentrenados (PLM) para evaluar su desempeño en la detección del humor y la ofensa. Para las tareas de regresión, promediamos las puntuaciones de diferentes modelos que conducen a un mejor desempeño que los modelos originales. Participamos en todas las subtareas. Nuestro sistema de mejor desempeño se clasificó 4 en la subtarea 1-b, 8 en la subtarea 1-c, 12 en la subtarea 2 y se desempeñó bien en la subtarea 1-a. Además, mostramos resultados completos utilizando diferentes modelos de lenguaje preentrenados que ayudarán como referencia para el trabajo futuro.

In this paper, we describe our system used for SemEval 2021 Task 7: HaHackathon: Detecting and Rating Humor and Offense.We used a simple fine-tuning approach using different Pre-trained Language Models (PLMs) to evaluate their performance for humor and offense detection.For regression tasks, we averaged the scores of different models leading to better performance than the original models.We participated in all SubTasks.Our best performing system was ranked 4 in SubTask 1-b, 8 in SubTask 1-c, 12 in SubTask 2, and performed well in SubTask 1-a.We further show comprehensive results using different pre-trained language models which will help as baselines for future work.

في هذه الورقة، نصف نظامنا المستخدم في SemEval 2021 المهمة 7: HaHackathon: الكشف عن الفكاهة والمخالفة وتقييمهما. استخدمنا نهجًا بسيطًا للضبط باستخدام نماذج لغوية مختلفة مدربة مسبقًا (PLMs) لتقييم أدائها للكشف عن الفكاهة والمخالفة. بالنسبة لمهام الانحدار، قمنا بحساب متوسط درجات النماذج المختلفة التي تؤدي إلى أداء أفضل من النماذج الأصلية. شاركنا في جميع المهام الفرعية. تم تصنيف نظامنا الأفضل أداءً في المرتبة 4 في المهمة الفرعية 1 -ب، و 8 في المهمة الفرعية 1 -ج، و 12 في المهمة الفرعية 2، وكان أداؤه جيدًا في المهمة الفرعية 1 -أ. كما أظهرنا نتائج شاملة باستخدام نماذج لغوية مختلفة مدربة مسبقًا والتي ستساعد كخطوط أساس للعمل المستقبلي.

Keywords

Artificial intelligence, Social Psychology, Economics, Humor Styles Questionnaire, Social Sciences, Epistemology, Psychological and Social Impact of Humor and Laughter, Task (project management), Artificial Intelligence, Machine learning, Automated Detection of Hate Speech and Offensive Language, FOS: Mathematics, Psychology, Natural language processing, Statistics, Computer science, Regression, Language model, FOS: Philosophy, ethics and religion, Management, FOS: Psychology, Detection, Philosophy, Computer Science, Physical Sciences, Simple (philosophy), Mathematics, SemEval

  • BIP!
    Impact byBIP!
    selected citations
    These citations are derived from selected sources.
    This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    0
    popularity
    This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
    Average
    influence
    This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
    Average
    impulse
    This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
    Average
Powered by OpenAIRE graph
Found an issue? Give us feedback
selected citations
These citations are derived from selected sources.
This is an alternative to the "Influence" indicator, which also reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Citations provided by BIP!
popularity
This indicator reflects the "current" impact/attention (the "hype") of an article in the research community at large, based on the underlying citation network.
BIP!Popularity provided by BIP!
influence
This indicator reflects the overall/total impact of an article in the research community at large, based on the underlying citation network (diachronically).
BIP!Influence provided by BIP!
impulse
This indicator reflects the initial momentum of an article directly after its publication, based on the underlying citation network.
BIP!Impulse provided by BIP!
0
Average
Average
Average
hybrid